Search

Background: Tubule cells can produce a variety of cytokines, extracellular matrix (ECM) components, and adhesion molecules in vitro and in vivo. It is generally assumed that stimulated tubule cells are proinflammatory and at least partially responsible for interstitial inflammation. However, the overall effect of tubular cells on interstitial cells is unknown. In this study, pro- and anti-inflammatory cytokine production and net effects on macrophages of tubule cells activated by lipopolysaccharide (LPS) were examined., Methods: Tubule cells stimulated with LPS expressed tumor necrosis factor-alpha (TNF-alpha), interleukin (IL)-1beta, IL-12, monocyte chemoattractant protein-1 (MCP-1), IL-10, and transforming growth factor-beta (TGF-beta). Conditioned media were collected from confluent monolayers of rat tubule cells stimulated, or not, by LPS for 4 and 18 hours, respectively. Macrophages were cultured with conditioned media and/or LPS (0.5 microg/mL) for 18 hours., Results: TNF-alpha and IL-lbeta mRNA of macrophages stimulated by LPS increased more than fivefold when cultured with control conditioned media from unstimulated tubule cells. Surprisingly, TNF-alpha and IL-lbeta levels of macrophages stimulated by LPS were not increased when cultured with conditioned media from activated tubule cells. Neutralizing antibodies to IL-10 and TGF-beta were used to define the inhibitory component(s) in conditioned medium. Anti-IL-10, but not anti-TGF-beta, abolished partially the inhibitory effects of conditioned media on macrophages., Conclusion: Tubule cells produce both pro- and anti-inflammatory cytokines and the net effect, partially explained by IL-10, of tubule cells activated with LPS is to inhibit activity of macrophages. Thus, the net effect of activated tubule cells on interstitial pathology may in certain circumstances, be anti- rather than pro-inflammatory.

We recently reported that inhibition of the transcription factor nuclear factor-kappaB (NFkappaB) with pyrrolidinedithiocarbamate (PDTC) reduced interstitial monocyte infiltration in rats with proteinuric tubulointerstitial disease, whereas N-acetylcysteine (NAC) was not effective. Here we investigate the effects of antioxidants (PDTC, NAC, and quercetin) on NFkappaB activation and cytokine transcription in primary cultured rat proximal tubular epithelial cells (PTC) stimulated with lipopolysaccharide. Antioxidant-mediated inhibition of NFkappaB activation (PDTC, 20-100 microM; NAC, 100 mM; and quercetin, 50 microM) diminished the induction of both pro- [interleukin (IL)-1beta, tumor necrosis factor-alpha, monocyte chemoattractant protein-1, macrophage inflammatory protein (MIP)-1alpha, and MIP-2] and anti-inflammatory (IL-10, transforming growth factor-beta1) cytokine transcription in PTC (RT-PCR analysis). PDTC and quercetin did not affect PTC viability, but NAC (100 mM) caused a threefold increase in lactate dehydrogenase leakage (P < 0.001). We conclude that NAC is unable to suppress NFkappaB activation in PTC at subtoxic and physiologically relevant concentrations. Furthermore, antioxidant-mediated inhibition of NFkappaB is correlated with the nonselective reduction of cytokine transcription in activated tubular cells. These data might explain the protective effects of PDTC-mediated NFkappaB inhibition in tubulointerstitial disease in vivo.

The transcription and translation of monocyte chemoattractant protein-1 (MCP-1), a CC chemokine, are increased in proximal tubule epithelial cells (PTC) stimulated with pathophysiologically relevant concentrations of albumin. The purpose of this study was to investigate whether nuclear factor kappaB (NFkappaB)/Rel proteins play a role in albumin-induced MCP-1 transcription. Confluent monolayers of rat PTC in primary culture were stimulated with delipidated bovine serum albumin. NFkappaB, the NFkappaB inhibitory protein (IkappaB), and MCP-1 transcription were assessed using electrophoretic mobility shift assays, Western immunoblotting, semiquantitative reverse transcription-PCR, and ribonuclease protection assays. Activation of NFkappaB by delipidated bovine serum albumin (15 mg/ml) was detectable within 2 h, maximal after 8 h, and maintained for at least 16 h of continuous exposure. Supershift analysis showed that the activated proteins were composed of p50/p50, p50/p65, and p50/c-Rel dimers. dimers. Cytoplasmic IkappaBalpha levels were decreased 30 min after stimulation and returned to unstimulated levels by 4 to 8 h. IkappaBbeta levels were decreased at 2 h and there was no recovery until 8 h. Inhibition of NFkappaB with pharmacologic agents (N-tosyl-phenylalanine chloromethyl ketone and dexamethasone) and an antisense oligonucleotide to the rat p65 subunit of NFkappaB significantly reduced MCP-1 transcription. The 3.6-kb 5' flanking region of the rat MCP-1 gene was cloned and sequenced, and two putative kappaB binding sites were identified within the enhancer region. Therefore, albumin increased NFkappaB and reduced IkappaB levels in PTC, and MCP-1 expression was dependent on NFkappaB activation. It is concluded that the activation of NFkappaB/Rel proteins modulates chemokine production in PTC in response to albumin and is likely to have an important role in the mediation of tubulointerstitial injury in proteinuric renal disease.

Cytokines play a pivotal role in synthesis and deposition of extracellular matrix in chronic renal failure (CRF). The proinflammatory properties of monocyte chemoattractant protein (MCP)-1 make it an ideal candidate cytokine for the production of interstitial inflammation in CRF. To investigate the possible role of proteinuria in inducing proximal tubular (PT) MCP-1, MCP-1 mRNA levels were measured by Northern blot and reverse transcription PCR in confluent monolayers of PT cells in primary culture in media containing a variety of proteins. PT cells produced MCP-1 mRNA in response to bovine serum albumin (BSA), delipidated BSA (dBSA; 0.5 to 30 mg/ml), holotransferrin, and apotransferrin (1 to 8 mg/ml). Unstimulated PT cells expressed very low levels of MCP-1 mRNA, detectable by reverse transcription PCR but not by Northern blot. The expression of MCP-1 mRNA reached a peak (sixfold greater than control) within 4 h of exposure to dBSA and was maintained for at least 24 h with continued exposure. Removal of dBSA from the media led to a rapid decline in MCP-1 mRNA expression. dBSA-induced MCP-1 expression was inhibited by lysine, an inhibitor of protein uptake, and reproduced by dBSA purified by gel and size-selective filtration. dBSA influenced MCP-1 expression at the level of transcription and probably translation, as evidenced by abrogation of MCP-1 by actinomycin D and superinduction with the protein synthesis inhibitor cycloheximide. The concentration of MCP-1 protein in response to dBSA added to the apical surface of PT cells was 2.4-fold greater in basolateral than in apical media, indicating basolateral secretion of MCP-1 protein. In summary, PT cell MCP-1 mRNA and protein expression are upregulated by albumin and transferrin, in concentrations similar to those of proteinuric urine. This effect could explain the link between proteinuria and interstitial inflammation in CRF.

The proposition that proximal tubule (PT) iron accumulation may cause PT injury by lysosomal destabilization or reactive oxygen species generation in human and animal chronic renal disease was examined in partially nephrectomized [remnant kidney (RK)] and sham-operated (SO) Wistar rats. Electron microscopic histochemistry with horseradish peroxidase indicated iron uptake into and release from lysosomes. PT cytoplasmic iron was seen in RK but not in SO by energy-dispersive X-ray spectrometry. Total (9.66 +/- 1.89 vs. 3.30 +/- 0.31 nmol/mg protein; P < 0.01), low-molecular-weight (1.39 +/- 0.09 vs. 0.91 +/- 0.07; P < 0.001), and catalytic iron (1.88 +/- 0.27 vs. 1.28 +/- 0.09; P = 0.05) were higher in RK cytoplasm than in SO. Lysosomal enzyme activity was greater in RK than in SO [e.g., N-acetyl-beta-D-glucosaminidase (NAG): 0.75 +/- 0.05 vs. 0.57 +/- 0.06 mumol p-nitrophenol.h-1.mg protein-1; P < 0.05] and was increased further by chronic iron loading (e.g., RK and NAG: 0.84 +/- 0.04 vs. 0.60 +/- 0.07; P < 0.05). There was no enzymatic evidence of lysosomal fragility, and chronic iron loading of RK decreased fragility as assessed by NAG release (1.36 +/- 0.14 vs. 2.17 +/- 0.14; P < 0.05).(ABSTRACT TRUNCATED AT 250 WORDS)

Iron accumulates within proximal tubular lysosomes in several models of renal disease and may play a role in the progression of proteinuric chronic renal disease by the generation of reactive oxygen species. In this study, tubular iron was examined at an ultrastructural level by energy dispersive x-ray spectrometry in streptozotocin (STZ) and BB diabetic rats, and in humans with diabetic nephropathy, and compared to their respective nondiabetic controls. Substantial amounts of iron accumulated in the secondary lysosomes of proximal tubules in STZ diabetic rats (4.16 +/- 0.47 iron-containing lysosomes/microns 2 x 10(-3) tubular area vs. 0.90 +/- 0.29 in controls, p < 0.001). Proximal tubular iron was related independently with urinary protein and transferrin excretion, suggesting increased cellular uptake of iron from the tubular fluid. Lysosomal iron accumulation was also associated with tubular damage (r = 0.55, p < 0.001). Minimal amounts of tubular iron were observed in BB diabetic and nondiabetic littermates. In humans with diabetic nephropathy, increased proximal tubular lysosomal iron concentration (35.6 +/- 13.0 mg% Fe vs. 9.5 +/- 2.7, p < 0.05) and numbers of iron-containing lysosomes were observed compared to nondiabetic controls, and the latter correlated with elevation of serum creatinine (r = 0.94, p = 0.016). These results suggest that filtered iron enters proximal tubular lysosomes across the brush-border membrane and are consistent with a role for iron in causing the tubular damage of diabetic nephropathy.