Tissue repair in rat kidney cortex after short treatment with aminoglycosides at low doses. A comparative biochemical and morphometric study.

The tissue repair subsequent to aminoglycoside-induced tubular necrosis has been evaluated for five different derivatives given at low doses. Gentamicin, dibekacin, netilmicin, and tobramycin were administered intraperitoneally three times a day at 10 mg/kg per day; amikacin was given intraperitoneally twice a day at 37.5 mg/kg per day. The drugs were administered for 4 or 10 days. Tissue regeneration in kidney cortex was assessed by: the rate of DNA synthesis (i.e., the specific radioactivity of DNA measured 1 hour after intraperitoneal injection of 200 microCi of [3H] thymidine per animal); the occurrence of less differentiated cells estimated by the morphometry of peroxisomes. In mature cells, identified by the presence of drug-induced myelin-like figures, peroxisomes occupied 3.0% of the cytoplasmic area, whereas in less differentiate cells devoid of myelin-like figures, the same organelles made 1.4% of the cytoplasmic area (mean values obtained from 20 treated animals). Gentamicin and amikacin were both shown to induce a dose-related enhancement of DNA synthesis in kidney cortex, except that the latter antibiotic was about 10 times less potent as compared to the former. When the five drugs were compared on the basis of their effect on DNA synthesis after 10 days of treatment, they gave the following ranking: gentamicin (5.6 times the mean control value) greater than or equal to dibekacin greater than netilmicin greater than tobramycin greater than amikacin (1.6 times the mean control value). The differences were less striking after 4 days of treatment. For each antibiotic, sections of proximal tubular cells obtained from four animals treated for 10 days (total surface, 2.10(4) microns2) were scanned for the presence of regenerative tissue (areas poor in peroxisomes). In gentamicin-treated rats, the areas poor in peroxisomes represented more than 18% of the scanned cytoplasmic surface in proximal tubular cells, as compared to about 4.5% in controls. The ranking of the five antibiotics according to the estimated area of undifferentiated tissue in proximal tubules was parallel to that found for DNA synthesis i.e., in gentamicin-treated rats undifferentiated tissue was much more prominent, as compared to amikacin-treated animals (about 6% of the scanned area). Since the relative extent of kidney cortex regeneration induced by each antibiotic at low dose is likely to reflect the degree of tubular necrosis, we conclude that the ranking outlined above indicates a difference in potential nephrotoxicity.