Renal globotriaosylceramide facilitates tubular albumin absorption and its inhibition protects against acute kidney injury.

To elucidate the physiologic function of renal globotriaosylceramide (Gb3/CD77), which up-to-date has been associated exclusively with Shiga toxin binding, we have analyzed renal function in Gb3-deficient mice. Gb3 synthase KO (Gb3S-/-) mice displayed an increased renal albumin and low molecular weight protein excretion compared to WT. Gb3 localized at the brush border and within vesicular structures in WT proximal tubules and has now been shown to be closely associated with the receptor complex megalin/cubilin and with albumin uptake. In two clinically relevant mouse models of acute kidney injury caused by myoglobin as seen in rhabdomyolysis and the aminoglycoside gentamicin, Gb3S-/- mice showed a preserved renal function and morphology, compared to WT. Pharmacologic inhibition of glucosylceramide-based glycosphingolipids, including Gb3, in WT mice corroborated the results of genetically Gb3-deficient mice. In conclusion, our data significantly advance the current knowledge on the physiologic and pathophysiologic role of Gb3 in proximal tubules, showing an involvement in the reabsorption of filtered albumin, myoglobin and the aminoglycoside gentamicin.

Comparing the potential renal protective activity of desferrioxamine B and the novel chelator desferrioxamine B-N-(3-hydroxyadamant-1-yl)carboxamide in a cell model of myoglobinuria.

Accumulating Mb (myoglobin) in the kidney following severe burns promotes oxidative damage and inflammation, which leads to acute renal failure. The potential for haem-iron to induce oxidative damage has prompted testing of iron chelators [e.g. DFOB (desferrioxamine B)] as renal protective agents. We compared the ability of DFOB and a DFOB-derivative {DFOB-AdAOH [DFOB-N-(3-hydroxyadamant-1-yl)carboxamide]} to protect renal epithelial cells from Mb insult. Loading kidney-tubule epithelial cells with dihydrorhodamine-123 before exposure to 100 µM Mb increased rhodamine-123 fluorescence relative to controls (absence of Mb), indicating increased oxidative stress. Extracellular Mb elicited a reorganization of the transferrin receptor as assessed by monitoring labelled transferrin uptake with flow cytometry and inverted fluorescence microscopy. Mb stimulated HO-1 (haem oxygenase-1), TNFα (tumour necrosis factor α), and both ICAM (intercellular adhesion molecule) and VCAM (vascular cell adhesion molecule) gene expression and inhibited epithelial monolayer permeability. Pre-treatment with DFOB or DFOB-AdAOH decreased Mb-mediated rhodamine-123 fluorescence, HO-1, ICAM and TNFα gene expression and restored monolayer permeability. MCP-1 (monocyte chemotactic protein 1) secretion increased in cells exposed to Mb-insult and this was abrogated by DFOB or DFOB-AdAOH. Cells treated with DFOB or DFOB-AdAOH alone showed no change in permeability, MCP-1 secretion or HO-1, TNFα, ICAM or VCAM gene expression. Similarly to DFOB, incubation of DFOB-AdAOH with Mb plus H2O2 yielded nitroxide radicals as detected by EPR spectroscopy, indicating a potential antioxidant activity in addition to metal chelation; Fe(III)-loaded DFOB-AdAOH showed no nitroxide radical formation. Overall, the chelators inhibited Mb-induced oxidative stress and inflammation and improved epithelial cell function. DFOB-AdAOH showed similar activity to DFOB, indicating that this novel low-toxicity chelator may protect the kidney after severe burns.

Influence of iron-overload on DNA damage and its repair in human leukocytes in vitro.

Iron is an important element that modulates the production of reactive oxygen species, which are thought to play a causative role in biological processes such as mutagenesis and carcinogenesis. The potential genotoxicity of dietary iron has been seldom studied in human leukocyte and only few reports have investigated in human colon tumor cells. Therefore, DNA damage and repair capacity of human leukocytes were examined using comet assay for screening the potential toxicity of various iron-overloads such as ferric-nitrilotriacetate (Fe-NTA), FeSO(4), hemoglobin and myoglobin, and compared with 200µM of H(2)O(2) and HNE. The iron-overloads tested were not cytotoxic in the range of 10-1000 microM by trypan blue exclusion assay. The exposure of leukocytes to Fe-NTA (500 and 1000 microM), FeSO(4) (250-1000 microM), hemoglobin (10 microM) and myoglobin (250 microM) for 30 min induced significantly higher DNA damage than NC. Treatment with 500 and 1000 microM of Fe-NTA showed a similar genotoxic effect to H(2)O(2), and a significant higher genotoxic effect than HNE. The genotoxicity of FeSO(4) (250-1000 microM), hemoglobin (10 microM) and myoglobin (250 microM) was not significantly different from that of H(2)O(2) and HNE. Iron-overloads generated DNA strand break were rejoined from the first 1h. Their genotoxic effect was not observed at 24h. These data from this study provide additional information on the genotoxicity of iron-overloads and self-repair capacity in human leukocytes.

Renoprotective effect of long acting thioredoxin by modulating oxidative stress and macrophage migration inhibitory factor against rhabdomyolysis-associated acute kidney injury.

Rhabdomyolysis-associated acute kidney injury (AKI) is a serious life-threatening condition. As such, more effective strategies are needed for its prevention. Thioredoxin-1 (Trx), a redox-active and macrophage migration inhibitory factor (MIF) modulating protein, has a short retention time in the blood. We examined the renoprotective effect of long acting Trx that was genetically fused with human serum albumin (HSA-Trx) against glycerol-induced AKI. An intravenous HSA-Trx pre-treatment attenuated the glycerol-induced decline in renal function, compared to a PBS, HSA or Trx alone. HSA-Trx caused a reduction in the tubular injuries and in the number of apoptosis-positive tubular cells. Renal superoxide, 8-hydroxy deoxyguanosine, nitrotyrosine and the plasma Cys34-cysteinylated albumin were clearly suppressed by the HSA-Trx treatment. Prior to decreasing TNF-α and IL-6, HSA-Trx suppressed an increase of plasma MIF level. In LLC-PK1 cells, HSA-Trx decreased the level of reactive oxygen species and lactate dehydrogenase release induced by myoglobin. HSA-Trx treatment resulted in a threefold increase in the survival of lethal glycerol-treated mice. The post-administration of HSA-Trx at 1 and 3 hr after glycerol injection exerted a significant renoprotective effect. These results suggest HSA-Trx has potential for use in the treatment of rhabdomyolysis-associated AKI via its extended effects of modulating oxidative stress and MIF.

Characterization of myoglobin toxicity in renal cortical slices from Fischer 344 rats.

Rhabdomyolysis is associated with acute renal failure. The following study first characterized myoglobin in vitro toxicity using renal cortical slices isolated from male Fischer 344 rats. This model provided interaction between various cells within the nephron and provides myoglobin access predominantly through the basolateral membrane. Second, this study examined the effect of deferoxamine (DFX) and glutathione on myoglobin toxicity to determine the role of radicals and iron. Renal cortical slices were incubated for 30-120 min with 0, 4, 10 or 12 mg/ml myoglobin. Myoglobin was pretreated with 4 mM ascorbic acid prior to addition to the slices to ensure that myoglobin was in its reduced state. In other experiments tissues were pretreated for 15 min with 0.1 mM of the iron chelator DFX or 30 min with 1 mM glutathione prior to co-incubation with myoglobin. Finally, slices were pretreated with 1 mM glutathione for 30 min, rinsed and incubated only with myoglobin. Early event changes occurred within a 60 min exposure and included a decline in pyruvate-stimulated gluconeogenesis, increased lipid peroxidation levels and decreased glutathione levels. Loss of ATP levels and increased lactate dehydrogenase (LDH) release required a 120 min exposure to myoglobin. DFX reduced myoglobin induced effects on LDH leakage but had no effect on gluconeogenesis suggesting that myoglobin toxicity had an iron dependent (LDH) and independent (gluconeogenesis) pathway. Pretreatment with glutathione provided complete protection and was mediated by intracellular events.

Characterization of myoglobin toxicity in renal cortical slices from Fischer 344 rats.

Rhabdomyolysis is associated with acute renal failure. The following study first characterized myoglobin in vitro toxicity using renal cortical slices isolated from male Fischer 344 rats. This model provided interaction between various cells within the nephron and provides myoglobin access predominantly through the basolateral membrane. Second, this study examined the effect of deferoxamine (DFX) and glutathione on myoglobin toxicity to determine the role of radicals and iron. Renal cortical slices were incubated for 30-120 min with 0, 4, 10 or 12 mg/ml myoglobin. Myoglobin was pretreated with 4 mM ascorbic acid prior to addition to the slices to ensure that myoglobin was in its reduced state. In other experiments tissues were pretreated for 15 min with 0.1 mM of the iron chelator DFX or 30 min with 1 mM glutathione prior to co-incubation with myoglobin. Finally, slices were pretreated with 1 mM glutathione for 30 min, rinsed and incubated only with myoglobin. Early event changes occurred within a 60 min exposure and included a decline in pyruvate-stimulated gluconeogenesis, increased lipid peroxidation levels and decreased glutathione levels. Loss of ATP levels and increased lactate dehydrogenase (LDH) release required a 120 min exposure to myoglobin. DFX reduced myoglobin induced effects on LDH leakage but had no effect on gluconeogenesis suggesting that myoglobin toxicity had an iron dependent (LDH) and independent (gluconeogenesis) pathway. Pretreatment with glutathione provided complete protection and was mediated by intracellular events.

Toxicity of myoglobin and haemoglobin: oxidative stress in patients with rhabdomyolysis and subarachnoid haemorrhage.

Haemolytic events, such as those following rhabdomyolysis and subarachnoid haemorrhage, often result in pathological complications such as vasoconstriction. Haem-protein cross-linked myoglobin and haemoglobin are generated by ferric-ferryl redox cycling, and thus can be used as markers of oxidative stress. We have found haem-protein cross-linked myoglobin in the urine of patients suffering from rhabdomyolysis and haem-protein cross-linked haemoglobin in the cerebrospinal fluid of patients following subarachnoid haemorrhage. These findings provide strong evidence that these respiratory haem proteins can be involved in powerful oxidation processes in vivo. We have previously proposed that these oxidation processes in rhabdomyolysis include the formation of potent vasoconstrictor molecules, generated by the myoglobin-catalysed oxidation of membranes, inducing nephrotoxicity and renal failure. Haem-protein cross-linked haemoglobin in cerebrospinal fluid suggests that a similar mechanism of lipid oxidation is present and that this may provide a mechanistic basis for the delayed vasospasm that follows subarachnoid haemorrhage.

Renal injury after muscle extract infusion in rats: absence of toxicity with myoglobin.

A crude muscle extract infused into rats produced oliguria, a precipitous drop in total hemolytic complement, and in circulating white cell and platelets counts. A mild vaso-depressor effects was noted. These changes were not produced by myoglobin or saline infusion. Muscle constituents other than myoglobin are responsible for the systemic and renal nephrotoxic effects observed.

Myoglobinuric acute renal failure in the rat: a role for medullary hypoperfusion, hypoxia, and tubular obstruction.

Myoglobin induces renal injury by mechanisms that remain incompletely defined. In this study, the effects of myoglobin upon renal microcirculation, oxygenation, morphology, and function were investigated in anesthetized rats, and the contribution of coexisting perturbations to myoglobin nephrotoxicity were evaluated. Myoglobin infusion (3.3 mg/min) reduced outer medullary blood flow and Po2, whereas renal blood flow and cortical Po2 were unaffected. Myoglobin infusion (38 mg/100 g weight over 45 min) induced renal failure associated with collecting duct and medullary thick ascending limb dilation and casts, with focal tubular damage, confined mainly to the superficial cortex. Preconditioning with indomethacin, I-N-monomethyl arginine, and theophylline reduced cortical superficial damage but enhanced injury within the inner stripe of the outer medulla and in medullary rays, the zones of lowest O2 supply. In preconditioned animals, tubulorrhexis was primarily observed in collecting ducts transversing the inner stripe, and was remarkably reminiscent of human descriptions (J. Oliver et al., J Clin Invest 1951; 30: 1307-1440). Deterioration in kidney function closely correlated with morphologic features of both tubular obstruction and necrosis. In conclusion, medullary vasoconstriction and intrarenal hypoxia may play a role in myoglobin-induced renal failure. The deterioration in kidney function appears to reflect the combined effects of cortical damage, medullary hypoxic injury, and tubular obstruction.

Intracellular myoglobin loading worsens H2O2-induced, but not hypoxia/reoxygenation-induced, in vitro proximal tubular injury.

Intracellular iron reportedly mediates many forms of tissue injury, including ischemic and myohemoglobinuric acute renal failure. This action may be explained by the ability of iron to catalyze the formation of the highly toxic hydroxyl radical (.OH) from H2O2 via the Fenton/Haber-Weiss reactions. To assess whether renal tubular myoglobin/iron loading, induced by a physiological mechanism (endocytosis), alters its susceptibility to O2 deprivation/reoxygenation- and H2O2-mediated injury, rats were infused with myoglobin or its vehicle (5% dextrose, control rats), and after 2 hours, proximal tubular segments (PTSs) were isolated for study. This infusion caused substantial myoglobin endocytic uptake (approximately 25 micrograms/mg PTS protein), and it doubled PTS catalytic iron content (assessed by bleomycin assay). Nevertheless, PTS viability (percent lactate dehydrogenase release) was minimally affected (4% to 6% increase), and an increased .OH burden (assessed by the salicylate trap method) did not appear to result. Deferoxamine addition, reported to protect against in vivo acute renal failure, paradoxically increased .OH levels (approximately 25%) in myoglobin-loaded, but not control, PTSs. Conversely, dimethylthiourea (an .OH scavenger) depressed .OH (by approximately 80%) in all PTSs. Myoglobin/iron loading modestly increased PTS vulnerability to exogenous H2O2 addition (P < .001). However, tubular susceptibility to hypoxia (15 and 30 minutes)/reoxygenation injury was not affected. .OH levels appeared to fall in response to both forms of injury, suggesting decreased .OH production and/or .OH scavenging. To assess whether myoglobin decreases .OH levels in the presence of Fenton reactants, myoglobin and six other test proteins were incubated with Fe2+/H2O2. Myoglobin decreased .OH levels by approximately 70%, a significantly greater decrement than was observed with the other proteins tested.(ABSTRACT TRUNCATED AT 250 WORDS)

Myoglobin inhibits proliferation of cultured human proximal tubular (HK-2) cells.

Following nephrotoxic injury, renal repair is dependent on tubular regeneration. In the case of myoglobinuric acute renal failure (ARF), persistence of myoglobin within tubular cells, or sublethal injury sustained at the height of exposure to it, might retard this process. To test this hypothesis, a human proximal tubular cell line (HK-2) was cultured for 24 hours in the absence or presence of clinically relevant myoglobin concentrations (0.5, 1, 2, 4 mg/ml). Immediately following myoglobin removal, lethal cell injury (vital dye uptake), lipid peroxidation, and DNA damage (alkaline unwinding assay) were assessed. The extent of cell proliferation was estimated over the next four days by a tetrazolium based (MTT) assay and by determining total intracellular LDH. Myoglobin's effects on protein and DNA synthesis were also assessed (35S-methionine and bromodeoxyuridine incorporation, respectively). Myoglobin induced dose-dependent lipid peroxidation (malondialdehyde generation) and cell death (up to 80% vital dye uptake with the 4 mg/ml challenge). Although 1 mg/ml myoglobin caused no cell death, it induced nearly complete growth arrest. This lasted for approximately three days following myoglobin removal from the media. Neither of two control proteins (albumin; lysozyme) nor a second nephrotoxin (gentamicin; 1 mg/ml) reproduced this effect. The 1 mg/ml myoglobin challenge caused an 80 to 90% depression in protein and DNA synthesis. It also induced significant DNA damage, as assessed by the alkaline unwinding assay (P < 0.01). Iron chelation therapy (deferoxamine) mitigated myoglobin-induced cell killing. However, its addition following myoglobin loading worsened HK-2 outgrowth by exerting a direct anti-proliferative effect. These results indicate that: (1) sublethal myoglobin toxicity can induce transient proximal tubular cell growth arrest, potentially slowing recovery from ARF; (2) this effect correlates with, and could result from, heme-induced DNA damage and a blockade in DNA/protein synthesis; and (3) deferoxamine can inhibit proximal tubular cell proliferation. This possibility needs to be considered in designing clinical trials with DFO for myohemoglobinuric ARF.

Antimicrobial peptides derived from heme-containing proteins: hemocidins.

Deprived of heme and partially unfolded hemoglobin, myoglobin and cytochrome c display microbicidal activity against a broad spectrum of microorganisms with half maximal lethal dose estimated at micromolar concentrations. The intact proteins were ineffective. Antibacterial activity of these apohemoproteins was also sustained after digestion to approximately 50 amino acids long peptides but further fragmentation abolished microbicidal properties. The most active fragment of apomyoglobin (corresponding to 56-131 region) showed a pronounced effect on the E. coli membrane permeabilization and its action was sensitive to salt as well as to divalent cations concentrations. The membrane-directed effect was specific toward bacteria but no lipopolysaccharide binding properties were observed. No hemolytic properties, even at high peptide concentrations were found; however, a slight but dose-independent cytotoxic effect was observed on fibroblasts and hepatoma cells. The presented data suggest a 'carpet-like' mechanism of the membrane-directed activity and may result from exceptional abilities of hemoprotein-derived peptides to form alpha-helical structures. We postulate that the antimicrobial peptides obtained from the heme-containing proteins should be named hemocidins, in contrast to, e.g., hemorphins displaying opioid-like activity.