点击显示 收起
【摘要】 Proteinuria is pathogenic to proximal tubular cells (PTC) and linked with progression to renal failure. The aim of this study was to determine the effects of human serum albumin (HSA) overload on the changes in gene and protein expression stimulated by oxidative stress in PTC and any interaction with ANG II that is pivotal in disease pathogenesis. Markers of oxidative stress, antioxidant defences, transcription factor activation, and the expression of stress-related genes were measured in human PTC (HK-2 cells) overloaded with either globulin-free fatty acid free (GF/FAF) HSA or globulin-free (GF) HSA. The effects of ANG II were also determined. HSA overload in HK-2 cells caused PTC hyperfunction, increased oxidative stress, and an upregulation of adaptive responses and stress-related genes. Some responses were common to both HSAs but others were unique to either HSA and unaffected by addition of ANG II or candesartan (a specific ANG II type 1 receptor blocker). ANG II also independently induced oxidative stress and upregulated other stress-related genes. HSA overload in HK-2 cells stimulated increased oxidative stress and upregulated changes in stress-related gene expression indicating new pathways of PTC injury that are not mediated via ANG II type 1 receptors.
【关键词】 stressrelated gene expression HK cells cell culture oxidative metabolism
IN DISEASED KIDNEYS PROTEINURIA and progression to end-stage renal failure are often linked in a wide variety of dissimilar disease states ( 1, 31, 42 ), and progression of the disease tends to correlate well with the extent of tubulointerstitial fibrosis (TIF) ( 6 ). A variety of studies have supported a direct pathogenic role of proteinuria in proximal tubular cells (PTC) ( 9, 38, 39, 44 - 47, 53 ). The mechanisms involved in the pathogenesis remain complex and multifactorial but include increased oxidative metabolism in the PTC triggering the generation of multiple mediators of tissue injury including chemokines, cytokines, and growth factors mediating TIF. It has been claimed that fatty acids (FA) bound to albumin rather than the protein per se mediate much of the pathogenic effects of proteinuria ( 30 ).
Reactive oxygen and nitrogen species (ROS) are important physiological regulators of redox processes and are increasingly implicated in the progression of different renal diseases ( 29, 39 ). Increased generation of ROS has been reported to lead to maladaptive responses to oxidative stress and/or in gene expression in PTC, but data in support of this are limited ( 28, 38, 39 ). Recent studies have indicated that exposure of PTC to albumin for 30 min increases the cellular hydrogen peroxide (H 2 O 2 ) content that serves as a signal for the activation of the ROS-sensitive nuclear transcription factor, NF- B (p50-65 subunits) ( 38 ).
ANG II is also independently involved in the pathogenesis of progressive renal injury in varied kidney diseases ( 20, 36 ). ANG II has complex injurious effects on the PTC, particularly in promoting the generation of mediators of TIF. These nonhemodynamic effects of ANG II are mediated primarily through ANG II type 1 receptors that are present throughout the kidney but expressed in increased numbers in the PTC ( 10 ). ANG II also appears to be a potent inducer of renal oxidative stress both in vivo and in vitro ( 16, 23, 24 ). Angiotensin-converting enzyme (ACE) inhibitors and ANG II receptor blockers reduce proteinuria and slow the rate of decline of renal function in patients with progressive kidney diseases ( 1, 31, 42 ), but it has not been possible to differentiate whether the benefit conferred by these agents is related to a reduction in proteinuria or to blocking proinflammatory, profibrotic, and other effects of ANG II ( 5 ).
This study addressed the potential involvement of oxidative stress in the responses of immortalized human PTC (HK-2) to overload with human serum albumin (HSA). Our hypothesis was that increased oxidative metabolism following HSA overload in PTC and the resulting increased oxidative stress would mediate the maladaptive responses and changes in gene expression. Markers of oxidative stress, antioxidant defences, transcription factor activation, and changes in stress-related gene expression were examined. The separate effects of ANG II on these processes were also determined.
METHODS AND MATERIALS
Cell Culture Techniques
Immortalized human proximal renal tubular epithelial HK-2 cells were obtained from the American Type Culture Collection (CRL-2190, Rockville, MD). Cells were tested and shown to be negative for mycoplasma using the Hoechst 33258 (Bisebenzamide) DNA staining method (Sigma, Poole, UK). Cells were cultured until 90-95% confluent into six-well plates or T75 flasks in DMEM/Ham's F12 medium supplemented with 5.5 mM glucose, 2 mM L -glutamine, 5 µg/ml insulin, 5 µg/ml transferrin, 5 ng/ml sodium selenite, 0.4 µg/ml hydrocortisone, 5 ng/ml epidermal growth factor, 100 U/ml penicillin, 100 µg/ml streptomycin, 20 mM HEPES, and 10% FCS.
Albumin Overload Model
Except where stated, the HK-2 cells were growth arrested for 48 h in serum-free medium (SFM) without added growth factors. Before harvest, cells were exposed for 24 h to one of two different HSA preparations (0-30 mg/ml); globulin-free fatty acid-free (GF/FAF) or globulin-free with attached fatty acids (GF) HSA (Sigma) in SFM. All studies were performed on cultures over a range of no more than 10 passages from the original stock.
Experiments with Exogenous ANG II and Candesartan
In some experiments, HK-2 cells were treated with either ANG II (1 µM) or candesartan (0.1 µM) or with both ANG II and candesartan for 24 h in SFM alone or supplemented with HSAs. Candesartan (AstraZeneca, Moelndal, Sweden) is a specific ANG II type 1 receptor blocker. In experiments in which ANG II and candesartan were used, HK-2 cells were first preincubated with candesartan in SFM for 1 h to saturate the ANG II receptor sites before the addition of ANG II.
Cell Viability Studies
Adherent and floating cells were harvested every 24 h for up to 72 h after exposure to either SFM alone (control) or to HSA preparations (0-30 mg/ml). Cell viability was assessed using the Trypan blue exclusion assay. The number of live and dead cells (stained blue) was counted, and the viability was expressed as the percentage of live cells within the total number of cells counted. Apoptosis was examined using terminal deoxynucleotidyl transferase biotin-dUTP nick end labeling (TUNEL) labeling in control cells and cells exposed for 24 h to HSA preparations (30 mg/ml), using TACS TdT apoptosis detection kit as recommended by the manufacturer (R&D Systems Europe).
Light Microscopy and Staining for Lipid Deposits
Oil red O staining was used to examine cellular lipid deposits in HK-2 cells grown on sterile glass coverslips and exposed for 24 h to SFM alone or SFM supplemented with HSAs. Cells fixed with 3.7% formaldehyde for 10 min at room temperature were stained for 10 min in 0.1% Oil red O in 60% isopropanol. The nuclei were counterstained with Harris hematoxylin, and the coverslips were mounted in aqueous mounting medium. Slides were examined by light microscopy (Zeiss Axiovert 200M, Carl Zeiss GmbH).
Endocytosis of Albumin by HK-2 Cells
This was determined using a modification of previously described methods ( 11, 34 ). Briefly, the HSA (either GF/FAF or GF HSA) preparations were labeled with fluorescein isothiocyanate (FITC) FluoroTag FITC conjugation kit (Sigma). Confluent cells were exposed for 45 min at 37 or 4°C to FITC-labeled albumin fractions (GF/FAF-FITC or GF-FITC) in SFM. Cells were lysed in 0.1% Triton X-100 in PBS. Samples were centrifuged and fluorescence was measured in the supernatants using FluoStar Optima microplate reader (BMG Labtechnologies). The protein content of the samples was determined by the BCA assay (Sigma).
Marker of Proximal Tubule Cell Turnover/Injury
Cells were exposed as described for 24 h either to SFM alone or SFM supplemented with 0-30 mg/ml of HSAs. The N -acetyl, D -glucosaminidase (NAG) activity released into the culture media was measured using a commercial kit (PPR Diagnostics, London, UK) based on the hydrolysis of 2-methoxy-4-(2'-nitrovinyl)-phenyl 2-acetamido-2-deoxy- D -glucopyranoside (MNP-GlcAc) ( 52 ).
Other Biochemical Analyses
Markers of oxidative stress.
GLUTATHIONE CONTENT. After incubation of HK-2 cells in SFM for 24 h in the presence or absence of HSAs, cells were harvested in 550 µl 1% sulphosalicylic acid (SSA; n = 6) and the samples were centrifuged at 11,000 g for 10 min at 4°C (Eppendorf, model 5402). The supernatants were collected and analyzed for glutathione (reduced and oxidized) using the glutathione recycling methods of Anderson ( 2 ) as modified by McArdle et al. ( 33 ). The reaction was followed at 415 nm using a Benchmark microplate reader (Bio-Rad). The protein pellets were washed in 1% SSA, and the protein content was determined using the modified Lowry method ( 33 ). Oxidized glutathione was measured using the same assay after derivitizing the samples with 2-vinylpyridine to prevent reduced glutathione from involvement in the cycling reaction.
TOTAL PROTEIN THIOL CONTENT. The pellets obtained following preparation with SSA in the glutathione assay were resuspended in 1 ml of ice-cold 0.5 M Tris·HCl buffer, pH 7.6, and the thiol content was determined using a 5,5-dithiolbis-2-nitro-benzoic acid methods ( 33 ). The absorbance was measured at 415 nm using a Benchmark microplate reader.
ASSESSMENT OF LIPID PEROXIDATION. After incubation for 24 h in the presence or absence of HSAs, cells were harvested in 550 µl ice-cold 2 M acetate buffer at pH 3.5 ( n = 6). The extent of lipid peroxidation was determined by measuring the malondialdehyde (MDA) content of HK-2 cells using HPLC with fluorescence detection ( 18 ). The protein content of the HK-2 cells was determined using a modified Lowry method ( 33 ).
ENZYME ACTIVITY AND PROTEIN ANALYSES
Analysis of antioxidant enzyme activities. Cells were harvested in 550 µl of PBS/well and sonicated for 10 s at 15 microns (Soniprep 150, Sanyo Scientific; n = 6). The cell extracts were centrifuged, and the protein content was measured by the BCA method (Sigma).
Catalase activity was measured using the method of Claiborne, based on the breakdown of H 2 O 2 ( 13 ). Glutathione peroxidase (GPx) activity was measured using the method of Flohe and Gunzler ( 17 ).
Superoxide dismutase activity. The superoxide dismutase (SOD) activity in HK-2 cells was assessed using activity gels as described by Beauchamp and Fridovich ( 4 ). Briefly, cells were harvested in 150 µl of 0.05 M potassium phosphate buffer containing 0.1 mM EDTA (pH 7.8). Cell extracts were then sonicated as described above and clarified by centrifugation (16,000 g for 10 min at 4°C). Ten micrograms of the total protein were resolved on 10% nondenaturing PAGE. The gels were stained in 4.9 mM nitro blue tetrazolium, 0.66 mM riboflavin, and 15 mM TEMED for 15 min in the dark. The activity of CuZnSOD and MnSOD was visualized under bright light as white bands on a dark blue background. The intensity of the signal from each band was corrected for background signal and expressed as a percentage of the activity of control cells and used to quantify enzyme activity (using Quantity One software, Bio-Rad, Hemel Hempstead, UK).
Western blotting. Control cells and cells exposed to HSAs ( n = 6) were harvested in 150 µl of 150 mM NaCl, 20 mM Tris·HCl, and 1% NP-40 (pH 7.5) supplemented with CompleteMini inhibitor cocktail (Roche Diagnostics). The cell suspensions were lysed on ice with intermediate vortexing. The samples were then centrifuged at 4°C for 5 min at 16,000 g, and the resulting supernatants were stored at -70°C. The protein concentration of the sample was determined using the BCA method. Aliquots of samples containing 35 µg of total protein were heated for 5 min at 95°C in denaturing loading buffer. The proteins were then resolved on 12.5% SDS-PAGE mini gels and blotted onto nitrocellulose membranes. The membranes were blocked overnight at 4°C in Blotto [5% skimmed milk in PBS supplemented with 0.05% Tween 20 (PBST)] and exposed for 1.5 h at room temperature to CuZnSOD and MnSOD primary antibodies (Stressgen) using antibody dilutions as recommended by the manufacturer. The membranes were repeatedly washed in PBST and exposed for 45 min at room temperature to peroxidase-conjugated secondary antibodies (1:10,000; Sigma). Protein bands were visualized using SuperSignal West Dura Extended duration chemiluminescent substrate (Pierce/Perbio Science) and recorded using a Chemidoc XRS imaging system and Quantity one software (Bio-Rad). The intensity of the signal from each band was corrected for background signal and expressed as a percentage of the content of control cells.
Molecular Biology Methods
Isolation of nuclear extracts. Cytoplasmic and nuclear extracts were isolated using the NE-PER Nuclear and Cytoplasmic Extraction Reagents Kit (Pierce/Perbio Science). Kit buffers were supplemented with a CompleteMini TM protease inhibitor cocktail (Roche Diagnostics) as recommended by the manufacturer. The protein content of HK-2 cells was determined by the BCA assay (Sigma).
EMSA. EMSA to examine DNA binding by the transcription factors NF- B and activator protein-1 (AP-1) was performed using Gel Shift Assay Systems (Promega) according to the manufacturer's protocols. Briefly, 2 µg of nuclear extract were incubated for 30 min on ice with [ 32 P]labeled NF- B or AP-1 consensus nucleotides. Samples were then analyzed on 6% nondenaturating acrylamide gel and the DNA-protein complexes were visualized using Storage Phosphor Screens (Amersham Pharmacia Biotech, Little Chalfont, UK) and Personal Molecular Imager FX (Bio-Rad). Further analyses of the images were carried out using Quantity one software (Bio-Rad).
In competition studies, 125-fold molar excess of cold NF- B or AP-1 oligonucleotides were incubated for 10 min at room temperature with the nuclear extract before the addition of [ 32 P]labeled NF- B or AP-1 nucleotides, respectively. SP-1 oligonucleotide (125-fold molar excess) was used as the nonspecific control.
RNA extraction. Total RNA was isolated from HK-2 cells grown in T75 flasks using Tri-reagent (Sigma) as recommended by the manufacturer's protocol. The RNA content of the samples was measured using RiboGreen RNA Quantitation Reagent (Molecular Probes, Invitrogen). The integrity of the total RNA was verified on denaturing agarose-formaldehyde gel.
cDNA array analysis. Atlas Human Stress Arrays with 234 stress-related genes immobilized on nylon membranes (Clontech, BD Biosciences) were used to identify the changes in mRNA expression in HK-2 cells in response to HSA overload with or without ANG II. To minimize the number of false positives and negatives in mRNA expression, equal amounts of total RNA from three separate experiments were pooled together and transcribed into [ 32 P]labeled first-strand cDNA using Atlas Array-specific primer mix. Labeled cDNA was purified from unincorporated 32 P by column chromatography, and the cDNA probes were allowed to hybridize to the atlas arrays for 16 h at 68°C using methods as recommended by the manufacturer. The arrays were then washed repeatedly and exposed for 6 and 72 h to Storage Phosphor Screens (Amersham Pharmacia Biotech). The phosphor screens were scanned using a Bio-Rad Personal Molecular Imager FX (Bio-Rad, Hercules, CA).
Real-time PCR. RT-PCR was performed with 1 µg of total RNA from HK-2 cells using iScript cDNA synthesis kit (Bio-Rad Laboratories) as recommended by the manufacturer. Real-time PCR was carried out using iCycler iQ Multicolor Real-Time PCR Detection System (Bio-Rad Laboratories), iQ SYBR Green Supermix (Bio-Rad Laboratories), and specific primers for human CREB1, extracellular SOD (EC-SOD), and MnSOD ( Table 1 ). Standard curves were generated with the primers by plotting copy numbers vs. threshold cycle (Ct). The linear correlation coefficient ( R 2 ) for the five primer pairs ranged between 0.996 and 0.999. The primer efficiencies were calculated based on the slopes of the standard curves ( E = 10 -1/slope ) ( 41 ). Real-time PCR reactions for each cDNA were performed simultaneously and normalized to GAPDH.
Table 1. Primer sequences used in real-time PCR analyses in HK-2 cells overloaded with albumin for 24 h
Data Analysis
Statistical analyses were carried out using the Statistical Package for Social Sciences (SPSS version 11.01, Surrey, UK). All data were expressed as means ± SE. Data were analyzed using a one-way repeated-measures ANOVA. Where a significant value was observed, Tukey's HSD post hoc analysis was performed to identify where the significant differences occurred. A P value of <0.05 was considered significant.
cDNA Array Data Analysis
The array images were analyzed using AtlasImage 2.01 bioinformatic software (Clontech, BD Biosciences). Global normalization was performed to compensate for different labeling efficiencies of RNA samples. To increase the stringency of the gene array data analysis further, a cut-off factor was calculated for each gene array as the average of the intensities of three weakly expressed genes (c- fos proto-oncogene, ERK-2, and XRCC2). This coefficient was subtracted from the intensity of each signal and 0 intensity was assigned to genes with corrected intensities of 0. Data presented represent the fold change in signal intensity compared with control cells. An increase exceeding twofold or a reduction to less than one-half in expression was considered significant.
Data Deposition
The gene array data contributing to this publication have been deposited in NCBIs Gene Expression Omnibus (GEO, http://www.ncbi.nlm.nih.gov/geo/ ) and are accessible through GEO Series accession number GSE5240.
RESULTS
Effect of Albumin Overload on Cell Viability, Morphology, and NAG Activity in HK-2 Cells
Cell viability. Cell viability assessed using Trypan blue was uniformly greater than 95% for both control (Co) cells incubated for up to 48 h with SFM alone or SFM supplemented with globulin-free fatty acid-free (GF/FAF) or globulin-free (GF) HSA (0-30 mg/ml). At 72 h, there was a significant decrease in cell viability for both Co cells and those exposed to HSA. Representative data are shown in Fig. 1 A from experiments using 30 mg/ml HSAs.
Fig. 1. Effect of human serum albumin overload on cell viability, morphology, and NAG activity in HK-2 cells. A : cell viability. i : HK-2 cells ( n = 6) were exposed for 24-72 h to SFM alone (Co), globulin-free fatty acid-free (GF/FAF) HSA, or globulin-free HSA with attached fatty acids (GF; 30 mg/ml HSA). * P < 0.05 cf. respective viabilities at 24 h. ii : HK-2 cells ( n = 6) were exposed for 24 h to GF/FAF HSA (30 mg/ml) in the presence or absence of ANG II (1 µM). * P < 0.05 cf. incubation with HSA alone. B : morphology. Cells were incubated for 24 h with SFM alone ( i ) or SFM supplemented with GF/FAF HSA ( ii ) or GF HSA ( iii ). Cells were stained with Oil Red O for lipid inclusions droplets, as indicated with arrows. C : extracellular NAG lysosomal activity. HK-2 cells ( n = 6) were exposed for 24 h to increasing concentrations of GF/FAF HSA (0-30 mg/ml). * P < 0.05 cf. control cells without HSA.
ANG II was found to decrease cell viability following 24-h exposure in both HSA-treated ( Fig. 1 A ) and Co cells (data not shown in detail). By 72 h posttreatment with ANG II, the loss of viability of cells treated with HSA alone was equivalent to ANG II and HSA-treated cells. Similar data occurred with both HSA preparations, but only data for GF/FAF HSA are presented.
In both Co cells and cells exposed to HSA incubation with candesartan alone or coincubation with candesartan and ANG II, there was no effect on cell viability regardless of the HSA fraction used (data not shown). There were also no differences in the rate of cell growth at 24, 48, or 72 h between cells exposed to HSA overload and those incubated with ANG II, candesartan, or coincubated with candesartan and ANG II (data not shown).
There was no evidence of significant apoptosis in either control cells or HK-2 cells exposed to HSA preparations at concentrations up to 30 mg/ml (data not shown).
HK-2 cell morphology. Lipid droplets were observed in the cytoplasm of HK-2 cells treated with GF HSA ( Fig. 1 B ). By contrast, there were no lipid inclusions seen in Co cells or in cells exposed to GF/FAF HSA. Representative images from cells treated with 30 mg/ml of GF/FAF and GF HSA are shown in Fig. 1 B ( ii and iii, respectively). ANG II treatment did not affect cell morphology in control cells or cells exposed to HSA (data not shown). There was no significant difference in the rate of uptake (endocytosis) of the FITC-labeled albumin with either HSA preparation (data not shown).
NAG activity. There was a significant increase in NAG release into the culture media following exposure of HK-2 cells to GF/FAF HSA compared with Co cells. NAG activity was significantly elevated following 0.5 mg/ml HSA but the increase plateaued over 2.5 mg/ml HSA ( Fig. 1 C ). With GF HSA, there was also an increase in NAG activity but this was only significant with greater than 5 mg/ml HSA (7.7 ± 0.6 µmol/mg) compared with Co cells (6.2 ± 0.1 µmol/mg).
Markers of Oxidative Stress in HK-2 Cells Overloaded with Albumin
Glutathione, protein thiol, and MDA content. There were no significant changes in the total glutathione and the total protein thiol content of HK-2 cells exposed to HSAs compared with Co cells regardless of the concentration of HSA used (data not shown). By contrast, exposure of HK-2 cells to either HSAs led to a significant increase in the percentage of glutathione in the oxidized form compared with Co cells, and this effect was greater using GF HSA fraction. This effect was apparent at greater than 2.5 mg/ml of GF HSA but did not increase further above 5 mg/ml HSA (representative data are shown in Fig. 2 A ).
Fig. 2. Markers of oxidative stress and adaptive responses in HK-2 cells overloaded with albumin for 24 h. A : oxidized glutathione (GSSG) content. HK-2 cells were exposed for 24 h to increasing concentrations of GF HSA (0-30 mg/ml; * P 0.05 cf. 0-30 mg/ml; # P 0.05 cf. 0 and 0.5 mg/ml). (Representative data from 3 experiments are shown.) B : MDA content. Effect of increasing concentrations of GF HSA (0-30 mg/ml) for 24 h on the MDA content of HK-2 cells (* P 0.05 cf. 0-5 mg/ml; # P 0.05 cf. 10, 15 mg/ml HSA). C : MnSOD activity. MnSOD activity was measured in Co cells and HK-2 cells were exposed to either GF/FAF HSA or GF HSA or to ANG II (1 µM; * P 0.05 cf. Co cells). Representative data from 3 independent experiments are shown.
Incubation of control cells with either ANG II or candesartan led to a significant reduction in the total glutathione content of HK-2 cells. Values fell from 53.7 ± 4.7 nmol/mg in Co cells to 38.9 ± 2.4 nmol/mg in cells after the addition of ANG II ( P < 0.05). There was a similar effect following incubation with candesartan alone [the total glutathione fell to 39.6 ± 1.7 nmol/mg ( P < 0.05)]. This effect was not seen following coincubation of cells with both ANG II and candesartan. The total glutathione content of HK-2 cells exposed to HSAs was unaffected by the addition of ANG II, candesartan, or coincubation with both ANG II and candesartan (data not shown). The oxidized glutathione content of control and HSA-treated cells was also unaffected by ANG II or candesartan addition neither was the thiol protein content of control or HSA-exposed cells following the addition of ANG II, candesartan, or both (data not shown in detail).
In titration experiments using GF HSA, the MDA content in HK-2 cells increased significantly in a dose-dependent manner above 10 mg/ml of GF HSA ( Fig. 2 B ) and was greater at 30 mg/ml (706.3 ± 36.9 pmol/mg). By contrast, the MDA content was significantly reduced in cells exposed to GF/FAF HSA compared with Co cells falling from 164.7 ± 19.4 to 104.0 ± 15.8 pmol/mg ( P < 0.05) on exposure to 30 mg/ml HSA (data not shown in detail).
The addition of ANG II significantly decreased the MDA content in cells exposed to SFM alone. Levels fell from 164.7 ± 19.4 to 88.5 ± 6.9 pmol/mg ( P < 0.05). Coincubation of Co cells with candesartan or with both ANG II and candesartan had no effect on the MDA content (data not shown). No significant changes were observed following exposure of cells to the different HSA preparations with ANG II, candesartan, or to both ANG II and candesartan.
Adaptation of HK-2 Cells to Oxidative Stress Induced by Albumin Overload
Catalase. Catalase activity was not altered in HK-2 cells incubated for 24 h with either HSA preparations compared with control cells. At 30 mg/ml HSA, catalase activity was 11.8 ± 1.3 m U/mg (GF/FAF HSA) and 11.2 ± 0.5 m U/mg (GF HSA) compared with control cells (13.0 ± 1.5 m U/mg). The addition of either ANG II or candesartan alone or coincubation with ANG II and candesartan also had no effect on catalase activity (data not shown).
Glutathione peroxidase. Glutathione peroxidase (GPx) activity was not significantly altered following HSA overload. At 30 mg/ml HSA, GPx activity was 346 ± 41 mU/mg (GF/FAF HSA) and 404 ± 28 mU/mg (GF HSA) compared with control cells (302 ± 30 mU/mg). The addition of either ANG II or candesartan alone or coincubation with ANG II and candesartan also had no effect on GPx activity (data not shown).
SOD. There was a significant increase in MnSOD activity in HK-2 cells exposed for 24 h to ANG II or to GF/FAF HSA but not following exposure to GF HSA ( Fig. 2 C ). By contrast, the activity of CuZnSOD was unaffected by exposure to ANG II or to either HSA preparation and was not significantly different to control cells throughout (data not shown).
Transcription Factor NF- B and AP-1 Activation in HK-2 Cells Overloaded with Albumin
NF- B. EMSA of nuclear extracts from HK-2 cells analyzed for NF- B DNA binding activity demonstrated the presence of three bands as indicated in Fig. 3 A. The nonspecific competitor (cold SP-1) decreased the intensity of band 1 but had no significant effect on bands 2 and 3. However the specific competitor (cold NF- B) reduced the intensity of bands 1 and 3 in comparison with the untreated extracts (positive control). ANG II or overload of the cells with either HSA had no effect on any of the three bands compared with untreated Co cells.
Fig. 3. EMSA of DNA binding by NF- B or AP-1 in nuclear extracts from HK-2 cells overloaded with albumin for 24 h. Nuclear extracts of cells exposed for 24 h to SFM alone (Co), ANG II (1 µM), or to HSAs (GF/FAF HSA or GF HSA) were subjected to EMSA for NF- B ( A ) or AP-1 ( B ). Unlabeled (cold) specific (NF- B or AP-1) or nonspecific (SP-1) oligonucleotides were used in 125-fold molar excess to confirm the specificity of the DNA-protein complexes. The complexes of labeled NF- B or AP-1 oligonucleotides are indicated with arrows. C : supershift analysis to examine the composition of AP-1. Nuclear extracts were incubated with antibodies raised against the human c-Jun and c-Fos proteins before incubation with [ 32 P]labeled AP-1 oligonucleotides. Cold specific (AP-1) or nonspecific (SP-1) oligonucleotides were used in 125-fold molar excess to confirm the specificity of the DNA-AP-1 complexes.
AP-1. Nuclear extracts from control HK-2 cells displayed a band representing AP-1 binding to DNA ( Fig. 3 B ). The specificity of the binding reaction for AP-1 was confirmed in a series of experiments with specific and nonspecific competitive inhibitors. Unlabeled (cold) AP-1 oligonucleotide (specific competitor) was able to reduce the intensity of the band, whereas there was no effect on this binding (band unaffected) using cold nonspecific competitor SP-1 ( Fig. 3 B ). A substantial increase of the AP-1 binding activity was seen following exposure of HK-2 cells to ANG II alone or to GF/FAF HSA and GF HSA (15 and 30 mg/ml). This effect was maximal after incubation with GF HSA as shown in Fig. 3 B. Also, following the addition of specific antibodies to c-Jun or c-Fos, there was a clear decrease of the AP-1 binding activity indicating that the complex represented a heterodimer of cJun/cFos ( Fig. 3 C ).
Expression of Stress-Related Genes in HK-2 Cells Overloaded with Albumin
Exposure of HK-2 cells for 24 h to different HSA preparations at 30 mg/ml modulated the expression of stress-related genes compared with Co cells cultured in SFM alone ( Table 2 ). The changes in gene expression are represented schematically in Fig. 4 A illustrating the number of upregulated or downregulated genes in cells treated with HSAs including those genes commonly affected by both HSAs. The majority of the genes affected by HSAs could be grouped into those involved in cell signaling and cell stress or antioxidant defences, protein turnover, and DNA repair/DNA replication, and these are listed in Table 3.
Table 2. Effect of albumin overload and ANG II on the expression of stress-related genes in HK-2 cells
Fig. 4. Stress-related gene profiling and protein content in HK-2 cells following ANG II or albumin overload for 24 h. A : Diagrammatic representation of the number of genes upregulated by exposure of cells to GF/FAF HSA or GF HSA. B : Diagrammatic representation of the number of genes upregulated by exposure of cells to GF/FAF HSA, GF HSA, or ANG II. C : Western blots for MnSOD and CuZnSOD. MnSOD and CuZnSOD content of HK-2 cells following exposure to ANG II, GF/FAF HSA, or to GF HSA. D : quantitative scanning densitometry of the Western blots for MnSOD ( i ) and CuZnSOD protein content ( ii ). Results are expressed as percentage of control (representative data from repeat experiments n = 3). * P 0.05 cf. control cells.
Table 3. Genes up- or down-regulated by GF/FAF HSA, GF HSA, and ANG II
Albumin overload in HK-2 cells using either HSA preparation (GF/FAF HSA or GF HSA) was associated with the upregulation of 29 stress-related genes by GF/FAF HSA and 26 genes with GF HSA. Of these upregulated genes 15 were commonly upregulated by both HSAs ( Fig. 4 A ). Additionally, exposure with either GF/FAF HSA or GF HSA (with attached fatty acids) led to further separate changes in 14 and 11 other genes being upregulated, respectively.
Incubation of HK-2 cells with ANG II alone induced upregulation in the expression of 48 stress-related genes ( Table 2 ). Almost half of these upregulated genes were unique to ANG II exposure and of these almost a third were related to heat shock protein (HSP) genes ( Tables 2 - 3 and Fig. 4 B ). Other genes affected exclusively by ANG II included those related to various transcription factors, genes involved in cell signaling, and DNA repair/DNA replication.
Moreover, out of the 15 genes commonly upregulated by both HSAs there were 10 genes that were also upregulated by ANG II including gene expression for the EC-SOD and transcription factors as illustrated in Fig. 4 B. Additionally, out of the 14 genes upregulated by GF/FAF HSA but not by GF HSA 11 were also upregulated after the addition of ANG II. The results of the gene array experiments also showed that out of 11 genes upregulated by GF HSA (but not by GF/FAF HSA) 6 were also commonly upregulated after the addition of ANG II.
Experiments to Examine the Validity of mRNA Data Obtained by Gene Array Analysis
Real-time PCR analysis of genes used to examine the validity of the cDNA array studies. Quantification of the mRNA levels of a number of genes shown to be increased by the cDNA array analyses was undertaken by real-time PCR. Of the three mRNA studied (CREB1, EC-SOD, and MnSOD), allwere found to be significantly increased by exposure to HSA in accord with data obtained from the cDNA arrays (data not shown in detail).
Analyses of protein changes in HK-2 cells overloaded with albumin for 24 h. As a further check on the validity of the cDNA array data, MnSOD and CuZnSOD proteins were analyzed. The protein levels of MnSOD were significantly increased in HK-2 cells exposed to ANG II and GF/FAF HSA ( Fig. 4 C ). By contrast, there was no substantial change in the CuZnSOD protein content of HK-2 cells after exposure to ANG II, GF/FAF HSA, or GF HSA. (The results of the activity of the MnSOD and CuZnSOD after exposure to HSAs and ANG II were previously described.)
DISCUSSION
We previously proposed a direct pathogenic role of protein, at the level of the PTC, in patients with diseased kidneys and proteinuria ( 47 ), and examined the validity of this hypothesis in a series of clinical studies including proteinuric patients with chronic allograft nephropathy ( 7, 44 - 47 ). We have now examined the potential effects of albumin overload (mimicking heavy clinical proteinuria) on oxidative stress in HK-2 cells and investigated the interplay between albumin overload and ANG II on the patterns of changes in cellular gene and protein expression stimulated by oxidative stress. ANG II has previously been reported to induce oxidative stress and to be pivotal in the pathogenesis of progressive kidney disease and fibrosis ( 16, 20, 23, 24, 36 ).
The results of the present study indicate that HSA overload causes 1 ) PTC hyperfunction, 2 ) oxidative stress, 3 ) adaptive responses, and 4 ) modification of stress-related gene expression in HK-2 cells. Some of these responses were common to both fatty acid-containing and fatty acid-free HSAs but others were unique to either the fatty acid-free albumin (GF/FAF) HSA or the fatty acid-containing HSA (GF). These data suggest that both albumin and the fatty acid moiety attached to albumin are important as separate factors in mediating responses but their pathogenic roles are not mutually exclusive with some features common to both. Thus the apparently disparate "proteinuria" hypotheses of mechanisms of injury are unlikely to be mutually exclusive. The effects of HSA overload were also independent of ANG II and unaffected by a specific ANG II type 1 receptor blocker, although there are ANG II type 1 receptors in HK-2 cells ( 48 ).
Effects of HSA Treatment on Cell Viability, Morphology, and PTC Hyperfunction in HK-2 Cells
In this model, there was increased PTC uptake of albumin, endocytosis, and catabolism following HSA treatment of HK-2 cells and lipid cellular accumulation following exposure to GF HSA rich in fatty acids, but HK-2 cells were viable at all concentrations of HSA. In contrast, exposure of cells to ANG II significantly reduced cell viability. Although differences in the isoelectric point of HSA preparations with or without attached fatty acids have been reported ( 15 ), this clearly had no effect on the endocytosis of the two HSA preparations used, making it unlikely that differences in the physical properties of the GF/FAF HSA and GF HSA fractions had contributed to their differential effects seen.
The prolonged albumin overload model was used to mimic the effects of chronic proteinuria, and we used different HSA preparations to dissociate and identify effects of the fatty acids attached to albumin from those of albumin alone. Cell culture models mimicking proteinuric conditions in patients are species specific and dependent on the concentrations of albumin used. Albumin has been reported to both promote PTC proliferation and hypertrophy (at lower concentrations 0-5 mg/ml) ( 26 ) but also induce 5-30 mg/ml in porcine PTCs) ( 14 ). Despite these data from other PTC lines that albumin leads to apoptosis ( 14, 49 ), we observed oxidative stress and adaptive responses with no evidence of decreased viability with even the highest level of HSA supplementation in HK-2 cells. This also emphasizes the importance of cell specificity on the effects of albumin overload in PTC culture systems.
Effect of HSA Overload on Oxidative Stress
There was a significant increase in oxidative stress in the HK-2 cells following HSA exposure shown by an increased oxidized glutathione and an increase in the content of the marker of lipid peroxidation MDA. These effects do not appear to have been previously reported. Control experiments excluded any artefactual effect of HSA exposure on MDA due to an increased generation of MDA by HSA per se or the presence of MDA as a contaminant of the GF HSA preparation. In contrast, HK-2 cells exposed to GF/FAF HSA or to ANG II showed a significant decrease of the MDA content compared with control cells. This was a consistent finding regardless of the concentration of HSA or of ANG II used (data not shown in detail). These data suggest a complex role for albumin and ANG II quite separate from pathogenic effects resulting in a reduction in lipid peroxidation by acting simultaneously through multiple cell survival pathways.
Adaptation of HSA Overloaded Cells to Oxidative Stress
Analyses of the activities of antioxidant enzymes suggest that GF/FAF HSA-treated HK-2 cells had adapted to the presence of increased superoxide radical within mitochondria by an increase in mitochondrial SOD (MnSOD). No changes were seen in the defences against cytosolic superoxide (CuZnSOD) or increased H 2 O 2 (catalase or glutathione peroxidise) production. Upregulation of the expression of the extracellular SOD gene (confirmed by real-time PCR) also followed exposure to both HSAs. These data have not to our knowledge been previously reported.
Thus our data appear to indicate that the mitochondria are the likely site for the increased ROS generation in PTCs with HSA overload, although there are a variety of other potential sources for increased ROS generation in PTCs ( 28, 38, 39 ). These include plasma membrane-associated sources of superoxide and H 2 O 2 [NAD(P)H oxidase and nitric oxide synthases], cytosolic sources (xanthine oxidase, cyclooxygenase, and lipooxygenase) in addition to mitochondria ( 19, 29, 50 ). ROS may initially modulate appropriate adaptive responses particularly in relation to cell signaling, synthetic pathways, and phagocytosis. However, with increased, uncontrolled, and sustained production, we postulate that they may become maladaptive facilitating aberrant activation of signaling pathways and transcription factors promoting TIF.
The redox-sensitive transcription NF- B and AP-1 factors are normally present in inactive forms in the cytoplasm of renal PTC. Activation and translocation to the nucleus follow a variety of stimuli including oxidative stress. We observed significant activation of AP-1, but not of NF- B, in HK-2 cells exposed for 24 h to either HSA preparations or to ANG II. This pattern of activation does not appear to have been previously reported, although others have shown activation of both NF- B ( 37, 38 ) and of AP-1 ( 37 ) in HSA-treated HK-2 cells. In this previous work, the period of exposure to HSA was briefer (30-60 min) and it may be that NF- B is activated for only a short period following exposure to HSA. The AP-1 transcription factor regulates many cellular processes and genes potentially involved in the progression of renal disease ( 35 ). The mechanisms involved in the induction of AP-1 activity are complex, but the mitogen-activated protein kinase (MAPKs) superfamily has an important role.
Changes in Stress-Related Gene Expression in HK-2 Cells Overloaded with Albumin
Significance of overall patterns of changes in gene expression. There were complex changes in gene expression linked with transcription factor activation and adaptive responses to oxidative stress, and we confirmed the validity of the DNA array data by examining some specific changes in gene expression in more detail. The novel data illustrate the complex pathogenic role of both the albumin and the fatty acid moiety attached to the albumin. Each appear to have independent effects in addition to triggering common changes in stress-related gene expression.
Commonly upregulated changes in gene expression by both HSA. Upregulation of important redox-sensitive cell signaling genes including ERK6 gene expression (p38 ) was observed in HSA-overloaded cells. p38 Is a recently identified member of the p38 family of MAPKs ( 32 ), which are particularly important in the signaling pathways promoting proliferation, inflammation, and fibrosis ( 43 ). Gene expression of ERK6 was upregulated following exposure to either HSA fraction or to ANG II.
Among the other commonly upregulated genes with both HSA preparations were those for two other important antioxidant enzymes: EC-SOD, the primary extracellular scavenger of superoxide in renal PTC, and the "gastrointestinal" cytoplasmic enzyme GPx (GI-GPx). EC-SOD is the least abundant of the SOD enzymes but is prominent in PTC ( 40 ).
Changes in gene expression specific to individual HSAs. The transcription factor CREB1 gene expression (a cAMP-responsive element-binding protein and a downstream target of ERK) was upregulated in the HK-2 cells (data confirmed by real-time PCR) following exposure to GF/FAF HSA (or ANG II) but was unaffected by GF HSA. CREB1 has multiple functions but with a particularly important role in cell survival in response to a moderate (nonlethal) ischemic/oxidative stress ( 3 ). These results suggest important differences between the two HSA in triggering different intracellular signaling pathways. Among other genes upregulated by GF/FAF HSA and ANG II, but not by GF HSA, were genes for MnSOD (confirmed by an increase in both MnSOD activity and MnSOD protein content). By contrast, GF HSA uniquely upregulated the ubiquitin-conjugating enzyme (UBE2A) mRNA levels. This is a part of the ubiquitin-protein ligase system involved in the transfer of activated ubiquitin moieties directly to a protein substrate ( 25 ), and these data are in keeping with a response to increased cellular protein turnover in HSA-treated cells.
ANG II-Induced Oxidative Stress and Changes in Gene Expression in HK-2 Cells Independently from Observed Effects of HSA
Markers of oxidative stress and adaptive responses. Treatment of HK-2 cells with ANG II for 24 h was found to be associated with increased oxidative stress and adaptive responses in agreement with other studies ( 16, 23, 24 ).
Changes in gene expression. Our results on gene array analysis following prolonged ANG II exposure are also novel and do not appear to have been previously reported. They illustrate complex regulation of gene expression by ANG II influencing regulatory processes linked with oxidative stress some of which are unique to the addition of ANG II and others commonly upregulated by one or both HSAs. Almost half of all genes upregulated following exposure to ANG II were unaffected by either HSA preparation, and a third of these were HSPs. These proteins can provide renoprotection via their chaperone functions or because of antioxidant effects ( 27 ). Other functions for HSP include blocking of apoptotic signaling pathways ( 12 ), which may have been relevant in the current model. In a previous study using rats, continuous ANG II infusion induced several HSPs in PTC, and some of these effects appeared to be independent of ANG II type 1 receptor activation ( 27 ). The expression of MAPKK1 was also upregulated only after exposure to ANG II, an effect that can potentially be linked to the role of ANG II as a profibrogenic cytokine leading to increased extracellular matrix protein deposition and fibrosis ( 36 ).
Previous data on the effects of ANG II using gene array analysis are limited. Braam et al. ( 8 ) used DNA microarrays analyses to study the direct effects of brief (4 h) exposure to ANG II (10 -7 -10 -8 M) in primary human PTC. They also reported ANG II induced rapid and complex regulation of the gene expression of several nuclear transcription factors and antioxidant genes (including EC-SOD as in our study).
Role of ANG II in HSA-treated HK-2 cells and pathophysiological implications. We anticipated that our data would show some synergy between the effects of ANG II and those of HSA overload on the measures of oxidative metabolism and oxidative stress but little was seen. We were unable to demonstrate that the addition of ANG II or the blocking of ANG II type 1 receptors with candesartan significantly affected markers of oxidative stress, adaptive responses, and transcription factor activation in the HSA-treated HK-2 cells. Rather our data suggested that the effects of HSA overload in HK-2 cells on these processes were independent of ANG II and not mediated via activation of ANG II type 1 receptor. We were also able to demonstrate that candesartan blocked the upregulation of more than two-thirds of the stress-related genes upregulated by ANG II in the HK-2 cells (data not shown) in keeping with the ability of candesartan to block angiotensin type 1 receptors in the HK-2 cells. In other studies, we showed that HK-2 cells express ANG II type 1 receptors ( 48 ). An alternative explanation for our data would be that the HSA overload masked the effects of ANG II at the concentrations used. However, the addition of ANG II to HK-2 cells led uniquely to the upregulation of 21 genes. If HSA had triggered the release of endogenous ANG II, one would have expected these genes to have been also upregulated after exposure to the HSA fractions. This was not seen and strongly argues against a simple masking effect by the HSA on the role of ANG II in the HK-2 cells.
Initiation of oxidative stress under conditions where ANG II is increased has been claimed to involve the activation of NADPH oxidase ( 22 ). However, the role of the different ANG II receptors in the generation of oxidative stress and the expression of NADPH oxidase in kidneys has not been established. Many of the pathogenic effects of ANG II are mediated by ANG II type 1 receptors, although ANG II type 2 receptors have been reported to have some separate pathogenic effects and also to have some effects that counter balance the effects of activation of ANG II type 1 ( 10, 20 ). However, these data have not been sustained in recent elegant rat experiments ( 51 ). Moreover, Remuzzi's group in mice ( 5 ) targeted deletion of the ANG II type 1A receptors and found that that this did not protect from progressive nephropathy following overload proteinuria suggesting that the toxic effect of protein on renal disease progression was not necessarily dependent on ANG II activation of type 1A receptors. Additionally, ANG II type 4 receptors have been described in HK-2 cells ( 21 ).
In conclusion, prolonged HSA overload in HK-2 cells was associated with complex effects in terms of oxidative stress, adaptive responses, and changes in stress-related gene expression. Some of these responses were similar with both HSAs, but others were unique to either GF/FAF HSA or GF HSA in keeping with important pathogenic roles for both the albumin and the fatty acid moiety and that their adverse effects are not mutually exclusive. These effects of HSA were also found to be independent of ANG II and not mediated via ANG II type 1 receptor activation. Thus these data suggest potential new pathways of PTC injury not linked with ANG II type 1 receptor activation or to hemodynamic mechanisms. Elucidation of these pathways potentially offers new approaches to intervention in an important and vexed problem that continues to drain resources.
GRANTS
This work was supported mainly by a grant from Mersey Kidney Research, Liverpool, UK with a contribution from the University of Liverpool Research Development Fund, UK and from Novartis Pharmaceutical UK.
【参考文献】
The GISEN Group (Gruppo Italiano di Studi Epidemiologici in Nefrologia. Randomised placebo-controlled trial of effect of ramipril on decline in glomerular filtration rate, and risk of terminal renal failure in proteinuric, nondiabetic nephropathy. Lancet 349: 1857-1863, 1997.
Anderson ME. Determination of glutathione and glutathione disulfide in biological samples. Methods Enzymol 113: 548-555, 1985.
Arany I, Megyesi JK, Reusch JE, Safirstein RL. CREB mediates ERK-induced survival of mouse renal tubular cells after oxidant stress. Kidney Int 68: 1573-1582, 2005.
Beauchamp C, Fridovich I. Superoxide dismutase: improved assays and an assay applicable to acrylamide gels. Anal Biochem 44: 276-287, 1971.
Benigni A, Corna D, Zoja C, Longaretti L, Gagliardini E, Perico N, Coffman TM, Remuzzi G. Targeted deletion of angiotensin II type 1A receptor does not protect mice from progressive nephropathy of overload proteinuria. J Am Soc Nephrol 15: 2666-2674, 2004.
Bohle A, Mackensen-Haen S, von Gise H, Grund KE, Wehrmann M, Batz C, Bogenschutz O, Schmitt H, Nagy J, Muller C, Muller G. The consequences of tubulointerstitial changes for renal function in glomerulopathies. A morphometric and cytological analysis. Pathol Res Pract 186: 135-144, 1990.
Bone JM, Amara AB, Shenkin A, Hammad A, Sells RA, Alexander JL, McArdle F, Rustom R. Calcineurin inhibitors and proximal renal tubular injury in renal transplant patients with proteinuria and chronic allograft nephropathy. Transplantation 79: 119-122, 2005.
Braam B, Allen P, Benes E, Koomans HA, Navar LG, Hammond T. Human proximal tubular cell responses to angiotensin II analyzed using DNA microarray. Eur J Pharmacol 464: 87-94, 2003.
Bruzzi I, Benigni A, Remuzzi G. Role of increased glomerular protein traffic in the progression of renal failure. Kidney Int Suppl 62: S29-S31, 1997.
Burnier M, Brunner HR. Angiotensin II receptor antagonists. Lancet 355: 637-645, 2000.
Caruso-Neves C, Kwon SH, Guggino WB. Albumin endocytosis in proximal tubule cells is modulated by angiotensin II through an AT2 receptor-mediated protein kinase B activation. Proc Natl Acad Sci USA 102: 17513-17518, 2005.
Chen Y, Currie RW. Heat shock treatment suppresses angiotensin II-induced SP-1 and AP-1 and stimulates Oct-1 DNA-binding activity in heart. Inflamm Res 54: 338-343, 2005.
Claiborne A. Catalase activity. In: CRC Handbook of Methods For Oxygen Radical Research. Boca Raton, FL: CRC, 1985, p. 283-284.
Erkan E, De Leon M, Devarajan P. Albumin overload induces apoptosis in LLC-PK 1 cells. Am J Physiol Renal Physiol 280: F1107-F1114, 2001.
Evenson MA, Deutsch HF. Influence of fatty acids on the isoelectric point properties of human serum albumin. Clin Chim Acta 89: 341-354, 1978.
Feliers D, Gorin Y, Ghosh-Choudhury G, Abboud HE, Kasinath BS. Angiotensin II stimulation of VEGF mRNA translation requires production of reactive oxygen species. Am J Physiol Renal Physiol 290: F927-F936, 2006.
Flohe L, Gunzler WA. Assays of glutathione peroxidase. Methods Enzymol 105: 114-121, 1984.
Fukunaga K, Suzuki T, Takama K. Highly sensitive high-performance liquid chromatography for the measurement of malondialdehyde in biological samples. J Chromatogr Sci 621: 77-81, 1993.
Geiszt M, Kopp JB, Varnai P, Leto TL. Identification of renox, an NAD(P)H oxidase in kidney. Proc Natl Acad Sci USA 97: 8010-8014, 2000.
Gesualdo L, Ranieri E, Monno R, Rossiello MR, Colucci M, Semeraro N, Grandaliano G, Schena FP, Ursi M, Cerullo G. Angiotensin IV stimulates plasminogen activator inhibitor-1 expression in proximal tubular epithelial cells. Kidney Int 56: 461-470, 1999.
Handa RK. Characterization and signaling of the AT( 4 ) receptor in human proximal tubule epithelial (HK-2) cells. J Am Soc Nephrol 12: 440-449, 2001.
Hannken T, Schroeder R, Stahl RA, Wolf G. Angiotensin II-mediated expression of p27Kip1 and induction of cellular hypertrophy in renal tubular cells depend on the generation of oxygen radicals. Kidney Int 54: 1923-1933, 1998.
Hannken T, Schroeder R, Zahner G, Stahl RA, Wolf G. Reactive oxygen species stimulate p44/42 mitogen-activated protein kinase and induce p27(Kip1): role in angiotensin II-mediated hypertrophy of proximal tubular cells. J Am Soc Nephrol 11: 1387-1397, 2000.
Haugen EN, Croatt AJ, Nath KA. Angiotensin II induces renal oxidant stress in vivo and heme oxygenase-1 in vivo and in vitro. Kidney Int 58: 144-152, 2000.
Hershko A, Heller H, Elias S, Ciechanover A. Components of ubiquitin-protein ligase system. Resolution, affinity purification, and role in protein breakdown. J Biol Chem 258: 8206-8214, 1983.
Iglesias J, Abernethy VE, Wang Z, Lieberthal W, Koh JS, Levine JS. Albumin is a major serum survival factor for renal tubular cells and macrophages through scavenging of ROS. Am J Physiol Renal Physiol 277: F711-F722, 1999.
Ishizaka N, Aizawa T, Ohno M, Usui Si S, Mori I, Tang SS, Ingelfinger JR, Kimura S, Nagai R. Regulation and localization of HSP70 and HSP25 in the kidney of rats undergoing long-term administration of angiotensin II. Hypertension 39: 122-128, 2002.
Ishola DA Jr, Post JA, van Timmeren MM, Bakker SJ, Goldschmeding R, Koomans HA, Braam B, Joles JA. Albumin-bound fatty acids induce mitochondrial oxidant stress and impair antioxidant responses in proximal tubular cells. Kidney Int 70: 724-731, 2006.
Jackson MJ, Papa S, Bolanos J, Bruckdorfer R, Carlsen H, Elliott RM, Flier J, Griffiths HR, Heales S, Holst B, Lorusso M, Lund E, Oivind Moskaug J, Moser U, Di Paola M, Polidori MC, Signorile A, Stahl W, Vina-Ribes J, Astley SB. Antioxidants, reactive oxygen and nitrogen species, gene induction and mitochondrial function. Mol Aspects Med 23: 209-285, 2002.
Kamijo A, Kimura K, Sugaya T, Yamanouchi M, Hase H, Kaneko T, Hirata Y, Goto A, Fujita T, Omata M. Urinary free fatty acids bound to albumin aggravate tubulointerstitial damage. Kidney Int 62: 1628-1637, 2002.
Lewis EJ, Hunsicker LG, Clarke WR, Berl T, Pohl MA, Lewis JB, Ritz E, Atkins RC, Rohde R, Raz I. Renoprotective effect of the angiotensin-receptor antagonist irbesartan in patients with nephropathy due to type 2 diabetes. N Engl J Med 345: 851-860, 2001.
Marinissen MJ, Chiariello M, Gutkind JS. Regulation of gene expression by the small GTPase Rho through the ERK6 (p38 gamma) MAP kinase pathway. Genes Dev 15: 535-553, 2001.
McArdle A, van der Meulen JH, Catapano M, Symons MC, Faulkner JA, Jackson MJ. Free radical activity following contraction-induced injury to the extensor digitorum longus muscles of rats. Free Radic Biol Med 26: 1085-1091, 1999.
McDonagh PF, Williams SK. The preparation and use of fluorescent-protein conjugates for microvascular research. Microvasc Res 27: 14-27, 1984.
Mezzano SA, Barria M, Droguett MA, Burgos ME, Ardiles LG, Flores C, Egido J. Tubular NF- B and AP-1 activation in human proteinuric renal disease. Kidney Int 60: 1366-1377, 2001.
Mezzano SA, Ruiz-Ortega M, Egido J. Angiotensin II and renal fibrosis. Hypertension 38: 635-638, 2001.
Morigi M, Buelli S, Angioletti S, Zanchi C, Longaretti L, Zoja C, Galbusera M, Gastoldi S, Mundel P, Remuzzi G, Benigni A. In response to protein load podocytes reorganize cytoskeleton and modulate endothelin-1 gene: implication for permselective dysfunction of chronic nephropathies. Am J Pathol 166: 1309-1320, 2005.
Morigi M, Macconi D, Zoja C, Donadelli R, Buelli S, Zanchi C, Ghilardi M, Remuzzi G. Protein overload-induced NF- B activation in proximal tubular cells requires H 2 O 2 through a PKC-dependent pathway. J Am Soc Nephrol 13: 1179-1189, 2002.
Nakajima H, Takenaka M, Kaimori JY, Hamano T, Iwatani H, Sugaya T, Ito T, Hori M, Imai E. Activation of the signal transducer and activator of transcription signaling pathway in renal proximal tubular cells by albumin. J Am Soc Nephrol 15: 276-285, 2004.
Nozik-Grayck E, Suliman HB, Piantadosi CA. Extracellular superoxide dismutase. Int J Biochem Cell Biol 37: 2466-2471, 2005.
Pfaffl MW. A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res 29: 2002-2007, 2001.
Remuzzi G, Benigni A, Remuzzi A. Mechanisms of progression and regression of renal lesions of chronic nephropathies and diabetes. J Clin Invest 116: 288-296, 2006.
Robinson MJ, Cobb MH. Mitogen-activated protein kinase pathways. Curr Opin Cell Biol 9: 180-186, 1997.
Rustom R, Grime JS, Costigan M, Maltby P, Hughes A, Shenkin A, Critchley M, Bone JM. Proximal renal tubular peptide catabolism, ammonia excretion and tubular injury in patients with proteinuria: before and after lisinopril. Clin Sci (Lond) 94: 425-430, 1998.
Rustom R, Grime JS, Maltby P, Stockdale HR, Critchley M, Bone JM. Observations on the early renal uptake and later tubular metabolism of radiolabeled aprotinin (Trasylol) in man: theoretical and practical considerations. Clin Sci (Lond) 84: 231-235, 1993.
Rustom R, Grime JS, Sells RA, Amara A, Jackson MJ, Shenkin A, Maltby P, Smith L, Hammad A, Brown M, Bone JM. Renal tubular peptide catabolism in chronic vascular rejection. Ren Fail 23: 517-531, 2001.
Rustom R, Jackson MJ, Critchley M, Bone JM. Tubular metabolism of aprotinin 99mTc and urinary ammonia: effects of proteinuria. Miner Electrolyte Metab 18: 108-112, 1992.
Shalamanova L, McArdle F, Jackson MJ, Amara AB, Bone JM, Rustom R. Protein overload and angiotensin II bolockade in proximal renal tubular cells: mechanisms of injury, metabolism and renoprotection. Renal Association/British Transplantation Society joint congress, The Waterfront Hall, Belfast, 2005, p. 398.
Tejera N, Gomez-Garre D, Lazaro A, Gallego-Delgado J, Alonso C, Blanco J, Ortiz A, Egido J. Persistent proteinuria upregulates angiotensin II type 2 receptor and induces apoptosis in proximal tubular cells. Am J Pathol 164: 1817-1826, 2004.
Vaziri ND, Dicus M, Ho ND, Boroujerdi-Rad L, Sindhu RK. Oxidative stress and dysregulation of superoxide dismutase and NADPH oxidase in renal insufficiency. Kidney Int 63: 179-185, 2003.
Wesseling S, Ishola DA Jr, Joles JA, Bluyssen HA, Koomans HA, Braam B. Resistance to oxidative stress by chronic infusion of angiotensin II in mouse kidney is not mediated by the AT 2 receptor. Am J Physiol Renal Physiol 288: F1191-F1200, 2005.
Yuen CT, Kind PR, Price RG, Praill PF, Richardson AC. Colorimetric assay for N -acetyl-beta- D -glucosaminidase (NAG) in pathological urine using the omega-nitrostyryl substrate: the development of a kit and the comparison of manual procedure with the automated fluorimetric method. Ann Clin Biochem 21: 295-300, 1984.
Zoja C, Donadelli R, Colleoni S, Figliuzzi M, Bonazzola S, Morigi M, Remuzzi G. Protein overload stimulates RANTES production by proximal tubular cells depending on NF-kappa B activation. Kidney Int 53: 1608-1615, 1998.
作者单位：Division of Metabolic and Cellular Medicine, School of Clinical Sciences, University of Liverpool, Liverpool, United Kingdom