Megalin contributes to kidney accumulation and nephrotoxicity of colistin

Abstract

Interest has recently been shown again in colistin because of the increased prevalence of infections caused by multidrug-resistant Gram-negative bacteria. Although the potential for nephrotoxicity is a major dose-limiting factor in colistin use, little is known about the mechanisms that underlie colistin-induced nephrotoxicity. In this study, we focused on an endocytosis receptor, megalin, that is expressed in renal proximal tubules, with the aim of clarifying the role of megalin in the kidney accumulation and nephrotoxicity of colistin. We examined the binding of colistin to megalin by using a vesicle assay. The kidney accumulation, urinary excretion, and concentrations in plasma of colistin in megalin-shedding rats were also evaluated. Furthermore, we examined the effect of megalin ligands and a microtubule-depolymerizing agent on colistin-induced nephrotoxicity. We found that cytochrome c, a typical megalin ligand, inhibited the binding of colistin to megalin competitively. In megalin-shedding rats, renal proximal tubule colistin accumulation was decreased (13.5 ± 1.6 and 21.3 ± 2.6 μg in megalin-shedding and control rats, respectively). Coadministration of colistin and cytochrome c or albumin fragments resulted in a significant decrease in urinary N-acetyl-β-d-glucosaminidase (NAG) excretion, a marker of renal tubular damage (717.1 ± 183.9 mU/day for colistin alone, 500.8 ± 102.4 mU/day for cytochrome c with colistin, and 406.7 ± 156.7 mU/day for albumin fragments with colistin). Moreover, coadministration of colistin and colchicine, a microtubule-depolymerizing agent, resulted in a significant decrease in urinary NAG excretion. In conclusion, our results indicate that colistin acts as a megalin ligand and that megalin plays a key role in the accumulation in the kidney and nephrotoxicity of colistin. Megalin ligands may be new targets for the prevention of colistin-induced nephrotoxicity.

Fig 1

Inhibitory effect of cytochrome c on the binding of colistin to megalin. Scatchard plots show the binding of colistin to megalin in the group given colistin alone and in the group given colistin with cytochrome c. BBMVs suspended in buffer A containing 0.1, 0.5, 1, 5, 10, or 100 μM colistin without or with 1 mM cytochrome c were incubated in a membrane suspension at 4°C for 60 min. Open circles represent a Scatchard plot of colistin alone, and solid circles indicate a Scatchard plot of colistin with cytochrome c. Each point represents the mean and SD of three measurements.

Fig 2

Effect of maleic acid on kidney cortex megalin expression. Western blotting was used to detect megalin in the renal cortexes of control rats and rats given an i.p. injection of maleate (400 mg/kg). Renal cortex samples (20 μg/lane) were subjected to 6% SDS-PAGE.

Fig 3

Kidney accumulation (A) and concentrations (B) of colistin in megalin-shedding rats. At 45 min after the administration of an i.p. injection of maleate (400 mg/kg) or saline (normal), colistin (0.5 mg/kg) was injected intravenously. Kidneys were collected at 180 min after the administration of colistin. Colistin contents in the kidneys (A) were examined by HPLC, and kidney colistin concentrations (B) were calculated by using kidney weight. Each column represents the mean and SD of four measurements. *, P < 0.05 (versus the control).

Fig 4

Cumulative amounts of colistin in the urine of megalin-shedding rats. The urinary bladders of rats were cannulated, and urine samples were collected at 60, 120, and 180 min after the administration of colistin. Open circles represent control rats, and solid circles represent megalin-shedding rats. Each point represents the mean and SD of four measurements. *, P < 0.05; **, P < 0.01 (versus the control).

Fig 5

Plasma colistin concentrations in megalin-shedding rats. The femoral arteries of rats were cannulated with polyethylene tubing, and blood samples were collected at 5, 15, 30, 45, 60, 90, 120, and 180 min after the injection of colistin. Open circles represent control rats, and solid circles represent megalin-shedding rats. Each point represents the mean and SD of four measurements.

Fig 6

Effects of megalin ligands on urinary NAG excretion. Rats were injected intravenously with saline (control), 1.0 mg/kg colistin alone, 1.0 mg/kg colistin with 100 mg/kg cytochrome c, or 1.0 mg/kg colistin with 50 mg FRALB. The rats were housed in metabolic cages to collect urine for 24 h. Each column represents the mean and SD of five measurements. *, P < 0.05 (versus the control); †, P < 0.05; ††, P < 0.01 (versus colistin alone).

Fig 7

Effects of megalin ligands on urinary colistin excretion. The 24-h urinary colistin excretion of groups treated with 1.0 mg/kg colistin alone, 1.0 mg/kg colistin with 100 mg/kg cytochrome c, and 1.0 mg/kg colistin with 50 mg FRALB is shown. The colistin contents of urine samples were examined by HPLC. Each column represents the mean and SD of five measurements. *, P < 0.05 (versus colistin alone).

Fig 8

Effect of colchicine on urinary NAG excretion. Rats were injected intravenously with saline (control), 1.0 mg/kg colistin alone, or 1.0 mg/kg colistin with 3.5 mg/kg colchicine. The rats were housed in metabolic cages to collect urine for 24 h. Each column represents the mean and SD of five measurements. *, P < 0.05 (versus the control); †, P < 0.05 (versus colistin alone).