Abstract
ABSTRACT. Abnormal traffic of proteins through the glomerular capillary has an intrinsic toxicity that results in tubular dysfunction and interstitial inflammation. It has been previously shown that in porcine proximal tubular cells high concentrations of albumin activated NF-κB, which is responsible for the enhanced synthesis of the inflammatory chemokine RANTES. This study investigates whether reactive oxygen species (ROS) served as second messengers in protein overload-induced NF-κB activation. Human proximal tubular cells (HK-2) were incubated (5 to 60 min) with human albumin and IgG (1 to 30 mg/ml). Both proteins induced a rapid or significant increase in hydrogen peroxide (H2O2) production at 5 min and persisting at 60 min. This effect was dose-dependent. The contribution of H2O2 in regulating NF-κB activation was evaluated by using the antioxidants dimethyl-thiourea and pyrrolidine dithiocarbamate in protein-overloaded HK-2 cells. Both agents, by preventing H2O2 generation, induced human albumin or IgG inhibited NF-κB activation. Stimulation of HK-2 with exogenous H2O2 resulted in the activation of a NF-κB subunit pattern similar to that obtained after protein challenge. Specific inhibitors of protein kinase C (PKC) activity significantly prevented H2O2 production and consequent NF-κB activation, suggesting that ROS generation in HK-2 cells occurs downstream of PKC activation. Either antioxidants or PKC inhibitor almost completely abolished the upregulation of the monocyte chemoattractant protein-1 gene induced by excess albumin, as evaluated by real-time PCR, thus supporting a role for PKC and ROS as critical signals for the expression of NF-κB-dependent inflammatory genes. To identify the enzymatic sources responsible for the increased H2O2 production, the effect of dyphenyleneiodonium, an inhibitor of the membrane NADP(H) oxidase, was studied, as was the effect of rotenone, which blocks complex I of the mitochondrial respiratory chain. It was found that both agents significantly reduced the exaggerated H2O2 induced by protein overload. These data indicate that exposure to excess proteins in proximal tubular cells induces the formation of ROS, which are responsible for NF-κB activation and consequent induction of NF-κB-dependent inflammatory signals.
Chronic renal diseases with highly enhanced glomerular permeability to proteins are accompanied by tubulointerstitial inflammation and scarring and progression to renal function deterioration (1). Experimental and human data in recent years have suggested that proteins filtered through the glomerular capillary barrier in excessive amount have an intrinsic renal toxicity linked to their over-reabsorption by proximal tubular cells and activation of tubular-dependent pathways of interstitial inflammation (1,2). Thus, in models of overload proteinuria, repeated injections of albumin in the rat increased glomerular barrier permeability and caused massive proteinuria and tubular changes with heavy macrophage and T lymphocyte infiltration into the renal interstitium (3). In rats with renal mass reduction, albumin and IgG accumulation by proximal tubular cells preceded the interstitial infiltration of inflammatory cells, which concentrated almost exclusively at the sites of proximal tubules congested with proteins (4).
Studies have been performed to define the biochemical pathways specifically activated by excessive tubular reabsorption, and evidence in vitro is now available that protein traffic induces proximal tubular cells to acquire an inflammatory phenotype (5). Thus, increasing concentrations of albumin in cultured rat proximal tubular cells upregulated the expression of monocyte chemoattractant protein-1 (MCP-1), one of the most powerful mononuclear cell attractants characterized so far (6). Similarly, albumin as well as IgG stimulated porcine proximal tubular cells to produce the chemotactic cytokine RANTES, secreted mainly toward the basolateral compartment (7), which would imply that in vivo RANTES generated in response to protein overload would accumulate in the interstitial space, thus contributing to inflammatory cell recruitment and subsequent fibrosis.
A candidate pathway of chemokine induction due to enhanced protein uptake is via NF-κB, a transcription factor of the Rel family comprising different members, including Rel A (p65), cRel, RelB, p50, and p52, which form homodimers or heterodimers with different affinities for variants of a decameric consensus binding site (8,9). The prototype NF-κB is composed of p50-p65 subunits. NF-κB exists in an inactive form in the cytoplasm of cells bound to the inhibitory protein IkB. NF-κB activation by appropriate triggers, such as cytokines, mitogens, viruses, and oxidants (8–10) promotes nuclear translocation of the DNA-binding subunits after they are released from IkB. We showed that albumin caused a dose-dependent increase in NF-κB activation in porcine proximal tubular cells in culture followed by the upregulation of RANTES, which was fully suppressed by NF-κB inhibitors (7).
Intracellular mechanisms leading to the activation of NF-κB after excess protein exposure have not been characterized so far. Evidence is available that reactive oxygen species (ROS) and particularly hydrogen peroxide (H2O2) can regulate NF-κB activation (11). Micromolar concentrations of H2O2 potently activated NF-κB in a human T cell line to a comparable or even higher extent than that achieved by phorbol myristate acetate and tumor necrosis factor (TNF) (12); this effect was prevented by the antioxidants n-acetyl-l-cysteine and pyrrolidine dithiocarbamate (PDTC) (12). The observation that different structurally unrelated antioxidants inhibited NF-κB activation induced by several stimuli other than H2O2 led to the hypothesis that ROS may be common messengers for various NF-κB-activating signals (11,13,14). As a corollary, it has been shown that many inducers of NF-κB can cause oxidative stress. Thus TNF-α and angiotensin II induced enhanced intracellular H2O2 production associated with NF-κB activation in endothelial cells (15,16).
A causal link between H2O2 production and NF-κB activation has been further confirmed by data obtained in cells overexpressing antioxidant enzymes with a well-defined activity in ROS metabolism. Cell lines stably transfected with catalase (17) as well as glutathione peroxidase (18), which degrades H2O2, were insensitive to NF-κB activation induced by TNF. By contrast, overexpression of cytosolic Cu/Zn superoxide dismutase by promoting cytosolic H2O2 accumulation potentiated this NF-κB response (17).
On the basis of the demonstration that renal tubular epithelial cells are capable of producing ROS after exposure to toxic products (19,20) or during hypoxia/reoxygenation (21), we investigated whether oxygen radicals served as messengers in protein overload-induced NF-κB activation in human proximal tubular cells. We also evaluated the potential role of protein kinase C (PKC) in mediating oxidant generation and NF-κB activation in this setting. Finally, we focused on the expression of MCP-1 as target gene of ROS-induced NF-κB activity in albumin-loaded tubular cells.
Materials and Methods
Cell Culture and Incubation
HK-2 cells are a permanent, well-characterized human proximal tubular cell line (22) and were obtained from the American Type Culture Collection (Rockville, MD, USA). They were grown in Dulbecco’s modified Eagle’s medium/F-12 with 5% fetal calf serum (FCS) supplemented with l-glutamine (2 mM), penicillin (100 U/ml), streptomycin (100 μg/ml), 3,3′,5-triiodo-l-thyronine (5 pg/ml), hydrocortisone (5 ng/ml), prostaglandin E1 (5 pg/ml), epidermal growth factor (10 ng/ml), insulin (5 μg/ml), transferrin (2.5 μg/ml), and sodium selenite (3.3 ng/ml). For experiments, cells were incubated in Hanks balanced salt solution buffer (HBSS).
To investigate the effect of protein overload on NF-κB activation, HK-2 cells were incubated for 30 min with buffer (control), 30 mg/ml of human serum albumin (HSA, Sigma Chemical Co., St. Louis, MO), or IgG (Sigma), and then electrophoretic mobility shift analysis (EMSA) and supershift of nuclear extracts were performed.
The involvement of ROS in NF-κB activation was assessed by the following methods. (1) We measured H2O2 production (supernatant and cells) in HK-2 cells exposed to buffer (control), HSA, or 30 mg/ml IgG for 5, 15, 30, and 60 min or to increasing concentrations of HSA or IgG (0, 1, 10, and 30 mg/ml) for 5 min. (2) We evaluated the effect of the antioxidants and metal chelators PDTC (100 μM; Sigma) (23) or 1,3-dimethyl-2-thiourea (DMTU [30 mM]; Sigma) (24,25) on both H2O2 production and NF-κB activation. Cells were treated with antioxidants 1 h before and during incubation with HSA or IgG (30 mg/ml) for 5 min. H2O2 levels were then measured in cell supernatant or intracellularly. NF-κB activation was assessed by EMSA in nuclear extracts of HK-2 cells pretreated (1 h) with the antioxidants and exposed to HSA or IgG (10 and 30 mg/ml) for 30 min in the presence of DMTU or PDTC. (3) We studied NF-κB activation (EMSA) in HK-2 cells exposed to exogenous H2O2 (5, 50, 100, and 500 μM; Sigma) for 30 min.
The following methods were used to study the role of PKC in ROS generation and NF-κB activation induced by protein overload. (1) H2O2 production was measured in HK-2 cells treated with a specific inhibitor of PKC, calphostin C (1 μM; Calbiochem-Novabiochemical Corporation, La Jolla, CA) (26,27) 1 h before and during incubation with HSA or IgG (30 mg/ml) for 5 min. (2) NF-κB activation was evaluated by EMSA in cells treated with the PKC inhibitors calphostin C or GF109203X (2.5 μM; Alexis Biochemical, San Diego, CA) 1 h before and during 30-min incubation with HSA or IgG (10 and 30 mg/ml).
To identify the enzyme(s) responsible for H2O2 production, proximal tubular cells were incubated with dyphenyleneiodonium (DPI 10 μM; Sigma), an inhibitor of the membrane NAD(P)H oxidase (28), or with rotenone (10 μM; Sigma), an inhibitor of complex I (29), 30 min before and during 5-min exposure to HSA or IgG (30 mg/ml).
Finally, to establish a functional link among H2O2 production, NF-κB activation, and NF-κB-dependent genes, we studied the expression of MCP-1 by real-time PCR in tubular cells loaded with HSA (30 mg/ml) for 6 h in the presence or absence of the antioxidants, DMTU and PDTC. Experiments with the PKC inhibitor calphostin C were also performed.
To exclude a possible toxic effect of the antioxidants, PKC inhibitors, DPI, or rotenone on proximal tubular cells, HK-2 cell viability was assessed by either trypan blue dye exclusion or tetrazolium-based colorimetric assay (MTT test). As reported in Table 1, tubular cell viability was not affected by any of the inhibitors.
Effect of human serum albumin (HSA) plus antioxidants, protein kinase C (PKC) inhibitors or dyphenyleneiodonium (DPI) and rotenone on tubular cell viabilitya
Preparation of Nuclear Extracts
Nuclear extracts were prepared from HK-2 cells by using the NE-PER Nuclear and Cytoplasmic Extraction Reagents Kit (Pierce/Celbio, Pero, Italy) according to the manufacturer’s instructions. To minimize proteolysis, all buffers contained a protease inhibitor cocktail (Protease inhibitor cocktail tablets; Roche Molecular Biochemicals, Sommerville, NJ). The protein concentration was determined by the Bradford assay by using the Bio-Rad (Richmond, CA) protein assay reagent.
Electrophoretic Mobility Shift and Supershift Assays
EMSA were performed as described previously (7) using the kB DNA sequence of the Ig gene (5′-CCGGTCAGAGGGGACTTTCCGAGACT). Nuclear extracts (2 μg) were incubated with 50 kcpm of 32P-labeled NF-κB oligonucleotide in a binding reaction mixture (10 mM Tris-HCl, pH 7.5, 80 mM NaCl, 1 mM ethylenediaminetetraacetic acid [EDTA], 1 mM dithiothreitol, 5% glycerol, 1.5 μg of poly(dI-dC)) for 30 min on ice. In competition studies, a 100-fold molar excess of unlabeled oligonucleotide was added to the binding reaction mixture as indicated before the addition of the labeled kB probe. For densitometric analysis, an equal-sized box was drawn around each band, and the volume density was determined in arbitrary units. The sum of the volume density of bands for a single sample was used as an indirect measure of NF-κB activation and expressed as a fold increase of the mean densitometry of respective control (represented as 1).
For supershift assays, the reaction mixture minus the probe was incubated for 1 h on ice with 1 μl of affinity-purified rabbit polyclonal antisera specific for p65 (sc-109), p50 (sc-114), RelB (sc-226), c-Rel (sc-71), and p52 (sc-298; Santa Cruz Biotechnology, Santa Cruz, CA). The labeled NF-κB oligonucleotide was then added, and the incubation was continued at room temperature for 20 min.
Determination of H2O2 Production
H2O2 generation was measured by the colorimetric Thurman assay (30). HK-2 cells were incubated with buffer (control), HSA, or IgG for appropriate time points in the presence of 1 mM azide. At the end of incubations, supernatants (1 ml) were collected and TCA 30% (wt/vol, 0.2 ml) was added. Cells were rinsed, detached, counted, washed, and resuspended in buffer (1 ml) before addition of TCA as above. After centrifugation, 10 mM ferrous ammonium sulfate (0.2 ml; Sigma) and subsequently 2.5 M potassium thiocyanate (0.1 ml; Sigma) were added to a 1.0-ml aliquot of the supernatant. The absorbance of the red ferrithiocyanate complexes formed in the presence of peroxides was measured at 480 nm and compared with a standard curve generated from dilutions of a reference solution of H2O2.
Intracellular H2O2 generation was also assessed in HK-2 cells exposed to HSA by the oxidant-sensitive dye, carboxy-dichlorodihydrofluorescein diacetate. Briefly, suspensions of HK-2 cells (1 × 106) were loaded with 5-(and 6)-carboxy-2′,7′-dichlorodihydrofluorescein diacetate (carboxy-H2DCFDA 5μM; Molecular Probes, Inc., Eugene, OR) for 10 min at 37°C. After centrifugation and washing to remove unincorporated probe, cells were incubated with buffer or HSA (30 mg/ml) at 37°C. For each sample, cellular fluorescence measurements were performed at 5, 15, 30, and 60 min by FACS (FACSort; Becton-Dickinson, San Jose, CA).
Real-Time PCR
Total RNA was extracted from HK-2 cells by the guanidium isothiocyanate/cesium chloride procedure as described previously (31). Contaminating genomic DNA was removed by RNase-free DNase (Promega, Ingelheim, Germany) for 1 h at 37°C. The purified RNA was reverse transcribed using random examers oligonucleotides and 200 U of SuperScript II RT (Life Technologies, San Giuliano Milanese, Italy) for 1 h at 42°C. No enzyme was added for reverse transcriptase-negative controls.
Real-time PCR was performed on TaqMan ABI 5700 Sequence Detection System (PE Biosystems, Warrington, UK) using heat-activated TaqDNA polymerase (Amplitaq Gold, PE Biosystems). The TaqMan PCR Reagent kit was used according to the manufacturer’s protocol. After an initial hold of 2 min at 50°C and 10 min at 95°C, the samples were cycled 40 times at 95°C for 15 s and 60°C for 60 s. Ct, or threshold cycle, is used for relative quantification of the input target number. The comparative Ct method normalizes the number of target gene copies to an endogenous control called reference, a housekeeping gene such as β-actin (ΔCt). Gene expression was then evaluated by the quantification of cDNA corresponding with the target gene relative to a calibrator sample serving as a physiologic reference (e.g., untreated cells, ΔΔCt). On the basis of exponential amplification of target gene as well as reference, the amount of amplified molecules at the threshold cycle is given by: 2−ΔΔCt.
The following oligonucleotide primers (300 nM) were used: hMCP-1: sense 5′-ACTCTCGCCTCCAGCATGAA, antisense 5′-GGGAATGAAGGTGGCTGCTA; β-actin: sense 5′-TCACCCACACTGTGCCCATCTACGA, antisense 5′-CAGCGGAACCGCTCATTGCCAATGG. All primers were obtained from Sigma Genosys (Cambridgeshire, UK).
Trypan Blue Uptake and MTT Test
To evaluate cell viability by trypan blue dye exclusion and the MTT test, HK-2 cells were treated with the antioxidants, DMTU (30 mM) or PDTC (100 μM), and the PKC inhibitors, calphostin C (1 μM) or GF109203X (2.5 μM), 1 h before and during 30-min incubation with HSA (30 mg/ml). DPI or rotenone were added 30 min before and during HSA exposure.
For trypan blue test, HK-2 cells were rinsed twice, detached with trypsin-EDTA, and resuspended in medium diluited 1:2 with trypan blue solution (Sigma). Live cells and trypan blue-stained dead cells were then counted by a hemocytometer chamber.
MTT test is a colorimetric method based on the cleavage of a yellow tetrazolium salt (3-(2-4) 2,5-diphenyl tetrazolium bromide) to purple formazan, which is directly proportional to the number of viable cells. For this procedure, HK-2 cells were washed twice with phosphate-buffered saline and exposed to 3 ml of MTT (0.5 mg/ml in DMEM without phenol red plus 10% FCS) for 1 h and 30 min. MTT solution was removed and replaced with 3 ml of extraction solution (90% isopropanol/10% DMSO; Sigma) for the same amount of time to solubilize the formazan crystals. The absorbance of each sample was measured by spectrophotometer at 570 nm with 630 nm as reference wavelength (32).
Statistical Analyses
Results are expressed as mean ± SE. Statistical analyses were performed using ANOVA followed by Tukey’s test for multiple comparisons. Statistical significance was defined as P < 0.05.
Results
Albumin and IgG Activate the Transcription Factor NF-κB in HK-2 Cells
The effect of HSA or IgG on NF-κB activation in human proximal tubular HK-2 cells is depicted in Figure 1. Nuclear extracts were assayed for activated NF-κB in an electrophoretic mobility shift assay of DNA binding factors using the radiolabeled consensus sequence kB probe containing the core kB site, GGGACTTTCC. Unstimulated HK-2 cells displayed two constitutive bands: an upper complex (complex I) and a faster-migrating lower complex (complex II). Thirty-minute incubation of proximal tubular cells either with HSA or IgG at the dose of 30 mg/ml induced a substantial rise in NF-κB binding activity of complexes I and II (Figure 1, A and B). The specificity of binding reaction was confirmed in competition experiments by the ability of excess unlabeled (cold) NF-κB oligonucleotide to inhibit binding.
Figure 1. Subunit composition of NF-κB activated by protein-overload in human proximal tubular cells. Electrophoretic mobility shift analyses (EMSA) were performed with nuclear extracts of human proximal tubular cells (HK-2) treated either with human serum albumin (HAS) or IgG (30 mg/ml) for 30 min. Complexes I and II denote the inducible kB-specific DNA-protein complexes. Nuclear extracts were incubated with antibodies against p65, p50, RelB, cRel, and p52 subunits. Antibody supershifts produced by binding of the antibody to the DNA-protein complex are indicated. To demonstrate the specificity of binding of the NF-κB oligonucleotide, a 100-fold molar excess unlabeled (cold) nucleotide was used to compete with the labeled NF-κB probe for binding to nuclear proteins. The results shown are representative of three independent experiments using three different preparations of nuclear extracts.
NF-κB complexes may constitute a variety of different homodimers and heterodimers; therefore, the subunit compositions of the protein overload-induced DNA complexes were analyzed by supershift assay. Rabbit polyclonal antisera specific for p65, p50, RelB, c-Rel, and p52 were added to nuclear extracts of HSA- or IgG-treated HK-2 cells. The results of both supershift analyses exibited the same subunit pattern. The upper band, complex I, consisted of p50/p65 heterodimer and p65/p65 homodimer because incubation of extracts with the anti-p65 antibody abolished the signal intensity of complex I, while anti-p50 antibody determined a signal reduction of the band. Antibodies to both subunits caused formation of more slowly migrating supershifted bands. Complex II was inhibited only by anti-p50, suggesting that the complex represented p50/p50 homodimer. Antisera to cRel, RelB, and p52 did not shift the NF-κB bands.
Effect of Protein Overload on H2O2 Production
To investigate whether protein overload-induced NF-κB activation was dependent on ROS generation, we studied the time course of H2O2 production in HK-2 cells exposed to HSA or IgG (30 mg/ml) (Figure 2). HSA induced a rapid release of H2O2 into the supernatant of tubular cells. A threefold increase over control of extracellularly released H2O2 was observed at 5 min, which was maintained thereafter. A parallel and significant increase in the intracellular H2O2 production occurred throughout the observation period. When HK-2 cells were exposed to IgG, both the extent and the time course of H2O2 production were fully comparable to those obtained in HSA-stimulated cells (Figure 2).
Figure 2. Time course of H2O2 generation in HK-2 cells exposed to HSA or IgG. HK-2 cells were exposed to buffer (control) or HSA or IgG (30 mg/ml) for 5, 15, 30, and 60 min. H2O2 production in cell supernatant or within the cells was assessed by colorimetric Thurman assay. Results are expressed as mean ± SE (n = 10). * P < 0.01 versus control at the corresponding time of incubation.
Increased production of H2O2 induced by HSA was further confirmed by FACS analyses of HK-2 cells preincubated with the oxidant-sensitive dye, DCFDA, used as an index of intracellular ROS generation. As shown in Table 2, the percentage of fluorescent cells significantly increased at 5 min after HSA stimulation, remaining elevated until 60 min.
Detection of intracellular H2O2 production in HK-2 cells exposed to HSA by the oxidant-sensitive dye, carboxy-H2DCFDAa
The effect of increasing concentrations of proteins on H2O2 production is shown in Figure 3. The amount of H2O2 released into the supernatant or found intracellularly in HK-2 cells exposed to HSA or IgG for 5 min was dose-dependent, reaching a statistically significant difference at 10 mg/ml.
Figure 3. Effect of increasing concentrations of HSA or IgG on H2O2 production by HK-2 cells. Cells were incubated with HSA (A and C) or IgG (B and D) at 0, 1, 10, or 30 mg/ml for 5 min. At the end of incubation, H2O2 production was measured in cell supernatants (A and B) or within the cells (C and D). Results are expressed as mean ± SE (n = 8). * P < 0.05 and ** P < 0.01 versus corresponding unstimulated control cells.
Protein Overload-Induced NF-κB Activation is H2O2-Dependent
We evaluated the effect of DMTU and PDTC, which readily enter the cells and serve as both oxidant scavengers and metal chelators, on ROS production induced by protein overload. Both agents significantly (P < 0.01) prevented the increase in extracellular and intracellular H2O2 induced by HSA or IgG (Table 3). DMTU and PDTC did not affect basal H2O2 production in supernatant (HK-2 + DMTU, 0.93 ± 0.26; HK-2 + PDTC, 1.01 ± 0.27; HK-2, 1.13 ± 0.13 nmol/106 cells) or cells (HK-2 + DMTU, 1.38 ± 0.24; HK-2 + PDTC, 1.11 ± 0.23; HK-2, 1.2 ± 0.21 nmol/106 cells).
Effect of antioxidants on H2O2 production in HK-2 cells exposed to protein overloada
To investigate whether H2O2 could function as second messenger in regulating NF-κB activation, we studied the effect of DMTU and PDTC on protein overload-induced NF-κB by EMSA experiments. As shown in Figure 4A, treatment of HK-2 for 30 min with increasing concentrations of HSA resulted in a significant (P < 0.01) activation of NF-κB already at 10 mg/ml. Pretreatment of cells with both antioxidants almost completely (P < 0.01) abolished the stimulatory effect of HSA. As shown in Figure 4B, IgG (10 and 30 mg/ml) induced NF-κB activation at an even greater extent than HSA, which was normalized (P < 0.01) by DMTU and PDTC.
Figure 4. Effect of the antioxidants, 1,3-dimethyl-2-thiourea (DMTU) and pyrrolidine dithiocarbamate (PDTC) on NF-κB activation induced by protein overload in HK-2 cells. (Top) EMSA were performed with nuclear extracts of HK-2 treated with DMTU (30 mM) or PDTC (100 μM) 1 h before and during 30-min incubation with HSA (A) or IgG (B) at 10 or 30 mg/ml. The results shown are representative of three independent experiments using three different preparations of nuclear extracts. (Bottom) Densitometric analyses of autoradiographic signals of NF-κB activity. Results are mean ± SE. # P < 0.01 versus control; * P < 0.01 versus HSA at the corresponding concentration. ° P < 0.01 versus IgG at the corresponding concentration.
Supershift analyses performed in nuclear extracts from tubular cells loaded with HSA in the presence of DMTU or PDTC showed that the antioxidants had the same effect on the subunit pattern of NF-κB activation consisting of p65/65 and p50/p50 homodimers and p65/p50 heterodimers.
To demonstrate a direct role of hydrogen peroxide on the activation of NF-κB, we exposed HK-2 to increasing concentrations of exogenous H2O2. Two bands with specific NF-κB binding activity were detected in nuclear extracts from HK-2 cells treated for 30 min with 5, 50, 100, and 500 μM H2O2 (Figure 5A). The lowest concentration of H2O2 approximates the amount produced by HSA-treated cells. As shown in Figure 5B, treatment with the antioxidants DMTU and PDTC markedly reduced NF-κB activation induced by H2O2. Supershift analyses revealed a subunit pattern of activation similar to that observed after protein overload (Figure 5C). In fact, addition of anti-p65 and anti-p50 antibodies resulted in a supershift of the upper band, represented by p50/p65 heterodimer and p65/p65 homodimer, and of the lower band, which consisted of p50/p50 homodimer.
Figure 5. Effect of exogenous H2O2 on NF-κB-DNA binding in HK-2 cells. (A) EMSA performed in nuclear extracts of HK-2 cells treated with buffer (control) or with H2O2 (5, 50, 100, 500 μM) for 30 min. (B) NF-κB-DNA binding activity analyzed in HK-2 cells treated with DMTU (30 mM) or with PDTC (100 μM) 1 h before and during 30-min incubation with H2O2 (500 μM). (C) Supershift assay of nuclear extracts of HK-2 cells treated with H2O2 (500 μM) for 30 min. The results are representative of three independent experiments using three different preparations of nuclear extracts.
Effect of PKC Inhibitors on H2O2 Generation and NF-κB Activation
It has been reported that PKC contributes in other cellular systems to the mediation of ROS-producing enzyme activation (33–35). Here we investigated whether PKC activation and ROS generation were critical sequential signals mediating protein overload-induced NF-κB activation in proximal tubular cells. Treatment of HK-2 cells with a specific PKC inhibitor, calphostin C, significantly reduced H2O2 production induced by HSA or IgG (30 mg/ml) either in the supernatant or within cells (Figure 6). H2O2 production in unstimulated cells did not change in the presence of calphostin C in supernatant (HK-2 + calphostin C, 1.13 ± 0.4; HK-2, 1.13 ± 0.13 nmol/106 cells) or cells (HK-2 + calphostin C, 1 ± 0.38; HK-2, 1.2 ± 0.21 nmol/106 cells).
Figure 6. Effect of calphostin C on protein overload-induced H2O2 generation. Cells were treated with the protein kinase C (PKC) inhibitor, calphostin C (1 μM), 1 h before and during incubation with HSA or IgG at 30 mg/ml for 5 min. H2O2 production was then measured in cell supernatant or within cells. Results are expressed as mean ± SE (n = 5). # P < 0.01 versus control; * P < 0.01 versus HSA; ° P < 0.05 and °° P < 0.01 versus IgG.
Blocking of ROS generation by calphostin C resulted in an almost complete inhibition of NF-κB activation induced by HSA and IgG (10 and 30 mg/ml), as documented by the densitometric analyses reported in Figure 7A. A similar reduction in NF-κB-DNA binding activity was observed when HSA-treated HK-2 cells were exposed to another PKC inhibitor, GF109203X (Figure 7B). Supershift analyses performed in nuclear extracts from tubular cells loaded with HSA in the presence of calphostin C showed a subunit pattern of NF-κB similar to that observed after antioxidant treatment.
Figure 7. Effect of PKC inhibitors on protein overload-induced NF-κB activation. (A) Densitometric analyses of autoradiographic signals of NF-κB activity in nuclear extracts from HK-2 exposed to calphostin C (1 μM) 1 h before and during incubation with HSA or IgG at 10 or 30 mg/ml for 30 min. Results are expressed as mean ± SE and are representative of three independent experiments using three different preparations of nuclear extracts. # P < 0.01 versus control; * P < 0.01 versus HSA at corresponding concentration; ° P < 0.05 and °° P < 0.01 versus IgG at corresponding concentration. (B) NF-κB-DNA binding activity in nuclear extracts from albumin-loaded HK-2 cells exposed to the PKC inhibitors, calphostin C and GF109203X (2.5 μM).
Enzymatic Source of Oxidant Generation
We attempted to identify the enzymes responsible for ROS production in response to protein overload. H2O2 production occurs very rapidly upon protein stimulation; therefore, we first tested the effect of DPI, an inhibitor of cell membrane NAD(P)H oxidase, and found that it reduced by about 60% either the amount of H2O2 released into the supernatant or found intracellularly upon cell stimulation with HSA or IgG (Figure 8A). We also looked at a possible involvement of the mitochondrial respiratory chain as a source of ROS in our experimental setting by using rotenone, an inhibitor of complex I. As shown in Figure 8B, this agent significantly (P < 0.01) inhibited H2O2 production in proximal tubular cells exposed to both proteins. DPI and rotenone did not affect basal H2O2 production in supernatant (HK-2 + DPI, 0.99 ± 0.44; HK-2 + rotenone, 1.05 ± 0.26; HK-2, 1.13 ± 0.13 nmol/106 cells) or cells (HK-2 + DPI, 0.94 ± 0.45; HK-2 + rotenone, 1.5 ± 0.38; HK-2, 1.2 ± 0.21 nmol/106 cells).
Figure 8. Effect of dyphenyleneiodonium (DPI) and rotenone on oxidant generation induced by HSA or IgG. Cells were exposed to DPI (A), an inhibitor of cell membrane-bound NAD(P)H oxidase or to rotenone (B), an inhibitor of complex I of the mitochondrial electron transport chain, 30 min before and during 5-min incubation with HSA or IgG at 30 mg/ml. H2O2 production was then evaluated in cell supernatants or within cells. Results are expressed as mean ± SE (n = 3). # P < 0.01 versus control; * P < 0.05 and ** P < 0.01 versus HSA; ° P < 0.05 and °° P < 0.01 versus IgG.
PKC-Induced Oxidant Generation Regulates NF-κB-Dependent MCP-1 Gene
To establish whether H2O2 production, NF-κB activation, and NF-κB-dependent genes could be functionally related, we studied the expression of MCP-1 by real-time PCR in tubular cells exposed to 30 mg/ml HSA for 6 h in the presence or absence of antioxidants. Albumin promoted a 3.8-fold increase in MCP-1 transcript levels as compared with control cells (Figure 9). DMTU and PDTC fully prevented MCP-1 gene upregulation, thus indicating the involvement of H2O2 in MCP-1 gene induction. On the basis of the above reported data that PKC regulates H2O2-dependent NF-κB activation, we also examined the effect of calphostin C on albumin-induced MCP-1 mRNA. As shown in Figure 9, MCP-1 gene overexpression was normalized by calphostin C treatment.
Figure 9. Effect of antioxidants and PKC inhibitors on albumin-induced monocyte chemoattractant protein-1 (MCP-1) gene expression by real-time PCR. Cells were treated with DMTU (30 mM), PDTC (100 μM), and calphostin C (1 μM) 1 h before and during 6-h incubation with 30 mg/ml HSA. The results shown are mean ± SE of three independent experiments. # P < 0.05 versus control; * P < 0.05 versus HSA.
Discussion
In this study, we have documented that albumin and IgG increased NF-κB-DNA binding activity in human proximal tubular cells and that the protein subunits involved were p65/p65 and p50/p50 homodimers and p50/p65 heterodimer. These data confirm our previous observation in porcine LLC-PK1 cells that albumin induced an intense NF-κB activation, which was responsible for the secretion of the chemokine, RANTES, across the basolateral cell membrane (7). As for the RANTES gene, other NF-κB-dependent genes, like MCP-1, were found to be rapidly induced in proximal tubular cells by high concentrations of proteins (6,36), which would lead to the assumption that overloading proximal tubular cells with various proteins results in a common pathway of activation of NF-κB-dependent genes that may possibly play a determining role in recruiting inflammatory cells into the renal interstitium. Relevant to this hypothesis are the in vivo evidences of renal NF-κB activation in a variety of experimental proteinuric nephropathies (37). In this context, we recently documented that in the rat models of 5/6 nephrectomy and passive Heymann nephritis, the progressive increase in urinary protein excretion was accompanied by a concomitant increase in renal NF-κB-DNA binding activity paralleled by upregulation of MCP-1 mRNA and interstitial infiltration of mononuclear cells (38).
How protein overload induces NF-κB-dependent genes in renal tubular cells remains to be elucidated. The purpose of this study was to investigate the intracellular signaling mechanism that underlies albumin- or IgG-induced NF-κB activation. Evidence that renal tubular epithelial cells are capable of producing oxidants (19-21), together with the observation in other cell systems that ROS activated NF-κB (15,16), prompted us to assess the effect of high concentrations of albumin or IgG on proximal tubular cell production of ROS as intermediates of NF-κB activation. We found that both albumin and IgG induced a dose-dependent and rapid increase in H2O2 production either in cell supernatant or at the intracellular level, which peaked at 5 min and persisted at 60 min. Treatment with the oxidant scavengers and metal chelators, DMTU and PDTC, prevented H2O2 generation and almost completely abolished the enhanced NF-κB activity induced by both proteins. An additional proof that H2O2 activated NF-κB in our experimental setting rests on the data that stimulation of HK-2 with increasing concentrations of exogenous H2O2 resulted in an NF-κB subunit pattern similar to that obtained after protein challenge. Actually, the lowest concentration of H2O2 (5 μM), which corresponds to the amount of hydrogen peroxide produced by protein overloaded tubular cells, already caused an increase in NF-κB activity.
Our findings demonstrate for the first time a functional link between protein overload-induced ROS generation and NF-κB activation in renal tubular cells. This is consistent with the in vitro demonstration in HeLa cells that internal stress caused by the accumulation of viral and cell membrane proteins in the endoplasmic reticulum induced NF-κB activation, which required both Ca++ and reactive oxygen intermediates as messengers (39).
To further elucidate the mechanisms leading to H2O2 generation and subsequent NF-κB activation, we focused on PKC, a family of ubiquitous phospholipid-dependent serine/threonine kinases, which plays a pivotal role in signal transduction implicated in cell differentiation, gene regulation, and proliferation (40). It is known that PKC translocates upon activation from the cytoplasm to cell membranes and mediates oxidant generation (34) and NF-κB activation in endothelial cells exposed to cytokines (33). Here we have documented that calphostin C, a specific PKC inhibitor, completely prevented H2O2 generation after 5-min exposure of tubular cells to albumin or IgG. Blocking of PKC-dependent ROS generation resulted in subsequent inhibition of the abnormal NF-κB-DNA binding activity observed at 30 min, as also confirmed by another PKC inhibitor, GF109203X. These findings strongly support a role of PKC as upstream regulator in the sequence of events leading to oxidant generation.
That PKC and oxygen radical generation induced by protein overload function as critical signals for the expression of NF-κB-dependent genes derives from data of real-time PCR experiments showing that either DMTU and PDTC or the calphostin C almost completely abolished the upregulation of the MCP-1 gene induced by albumin in proximal tubular cells.
Finally, to identify the enzymes responsible for H2O2 production induced by protein overload, we first looked at the membrane-bound multicomponent enzyme complex, NAD(P)H oxidase, on the basis of evidence that (1) activation of the membrane-bound NAD(P)H oxidase promoted superoxide anion (O2) generation in renal proximal tubular cells exposed to angiotensin II, probably through an increase in p22-phox transcript, one of the key electron transfer elements of the NAD(P)H oxidase complex (41); (2) in cultured vascular cells, the production of ROS induced by high glucose levels occurred through a PKC-dependent activation of NAD(P)H oxidase (34), and in phagocytic cells, PKC also regulated the phosphorylation of a cytosolic NAD(P)H oxidase component, p47-phox, which is critical for the enzyme activation (42). We found that the NAD(P)H oxidase inhibitor, DPI, reduced albumin and IgG-induced H2O2 production in HK-2 cells, suggesting that NAD(P)H oxidase represents an enzymatic source of intracellular H2O2 generated in response to protein overload. It is tempting to speculate that binding of albumin or IgG to their receptors (43–45) leads to a rapid membrane-bound PKC-dependent activation of NAD(P)H oxidase with generation of O2⨪, which dismutes to H2O2. The fact that PKC blockade by calphostin C prevented the excessive H2O2 production upon protein challenge would imply a role of PKC in the activation of NAD(P)H oxidase in our experimental setting.
Activation of NF-κB induced by several stimuli, including H2O2, requires an intact mitochondrial electron transport chain (46,47); therefore, we have also investigated a possible contribution of mitochondria to H2O2 generation in protein-overloaded HK-2 cells. Mitochondria are a major source of ROS production, and two main sites of O2⨪ and H2O2 generation in the inner mitochondrial membrane have been identified: NADH dehydrogenase at complex I, and ubiquinone at complex III (48). Our data showing that rotenone significantly reduced albumin and IgG-induced H2O2 production by blocking the electron flow from NADH dehydrogenase to ubiquinone suggest that the mitochondrial electron transport chain is also a source of intracellular H2O2 in proximal tubular cells in response to protein overload.
In summary, we have shown that in human proximal tubular cells (1) albumin and IgG dose-dependently elicited a rapid and sustained H2O2 generation over time; (2) H2O2 served as a signal for NF-κB activation as indicated by experiments with the antioxidants, DMTU and PDTC, and that preventing H2O2 formation reduced NF-κB activation in response to protein overload; (3) PKC appeared as an upstream regulator in the sequence of events leading to oxidant-induced NF-κB activation; (4) PKC-dependent ROS generation induced by albumin was instrumental for the expression of the NF-κB-dependent MCP-1 gene; and (5) membrane NAD(P)H oxidase and mitochondria were required for H2O2 generation.
These findings suggest that tubular protein congestion induces a PKC-dependent oxidant generation responsible for enhanced NF-κB activity and consequent induction of NF-κB-dependent pathways of interstitial inflammation.
Acknowledgments
Part of this work was presented at the 32nd Annual Meeting of the American Society of Nephrology, November 1 to 8, 1999, Miami Beach, Florida. Dr. Simona Buelli is a recipient of a fellowship “In memory of Angelo Battista Pedrali.” We are indebted to Professor Schlondorff and Dr. Peter Nelson for their invaluable help in setting up the EMSA method.
Footnotes
-
Dr. Bruce Kone served as Guest Editor and supervised the review and final disposition of this manuscript.
- © 2002 American Society of Nephrology
References
Jump to section
More in this TOC Section
Cited By...
- Simultaneous activation of innate and adaptive immunity participates in the development of renal injury in a model of heavy proteinuria
- Kidney Disease Measures and Left Ventricular Structure and Function: The Atherosclerosis Risk in Communities Study
- Megalin/Cubulin-Lysosome-mediated Albumin Reabsorption Is Involved in the Tubular Cell Activation of NLRP3 Inflammasome and Tubulointerstitial Inflammation
- Albuminuria Is an Appropriate Therapeutic Target in Patients with CKD: The Pro View
- Glutathionylation of the Aquaporin-2 Water Channel: A NOVEL POST-TRANSLATIONAL MODIFICATION MODULATED BY THE OXIDATIVE STRESS
- Association of Urinary Injury Biomarkers with Mortality and Cardiovascular Events
- (Na+ + K+)-ATPase Is a Target for Phosphoinositide 3-Kinase/Protein Kinase B and Protein Kinase C Pathways Triggered by Albumin
- Luminal Alkalinization Attenuates Proteinuria-Induced Oxidative Damage in Proximal Tubular Cells
- Hydrogen and N-Acetyl-L-Cysteine Rescue Oxidative Stress-Induced Angiogenesis in a Mouse Corneal Alkali-Burn Model
- Immunoglobulin Light Chains Activate Tubular Epithelial Cells through Redox Signaling
- PKC-{delta} Promotes Renal Tubular Cell Apoptosis Associated with Proteinuria
- Hypertonicity-Induced Expression of Monocyte Chemoattractant Protein-1 through a Novel Cis-Acting Element and MAPK Signaling Pathways
- The Regulation of NADPH Oxidase and Its Association with Cell Proliferation in Human Lens Epithelial Cells
- Remission of Proteinuria Improves Prognosis in IgA Nephropathy
- PPAR{alpha} Protects Proximal Tubular Cells from Acute Fatty Acid Toxicity
- Immunoglobulin Light Chains Generate Hydrogen Peroxide
- Mitochondria Are the Major Targets in Albumin-Induced Apoptosis in Proximal Tubule Cells
- Oxidants in Chronic Kidney Disease
- How Does Proteinuria Cause Progressive Renal Damage?
- Intracellular Delivery of the p38 Mitogen-Activated Protein Kinase Inhibitor SB202190 [4-(4-Fluorophenyl)-2-(4-hydroxyphenyl)-5-(4-pyridyl)1H-imidazole] in Renal Tubular Cells: A Novel Strategy to Treat Renal Fibrosis
- Transcriptional Regulation of Nephrin Gene by Peroxisome Proliferator-Activated Receptor-{gamma} Agonist: Molecular Mechanism of the Antiproteinuric Effect of Pioglitazone
- Rapamycin Ameliorates Proteinuria-Associated Tubulointerstitial Inflammation and Fibrosis in Experimental Membranous Nephropathy
- Albumin Activates ERK Via EGF Receptor in Human Renal Epithelial Cells
- Induction of Renal Tubular Cell Apoptosis in Focal Segmental Glomerulosclerosis: Roles of Proteinuria and Fas-Dependent Pathways
- Activation of the Signal Transducer and Activator of Transcription Signaling Pathway in Renal Proximal Tubular Cells by Albumin
- Protein Overload Induces Fractalkine Upregulation in Proximal Tubular Cells through Nuclear Factor {kappa}B- and p38 Mitogen-Activated Protein Kinase-Dependent Pathways
- Anemia Management and the Delay of Chronic Renal Failure Progression
- Imbalance of T-Cell Subsets in Angiotensin II-Infused Hypertensive Rats With Kidney Injury
- Proteinuria and Phenotypic Change of Proximal Tubular Cells