Gentamicin-induced apoptosis in renal cell lines and embryonic rat fibroblasts.

Gentamicin, an aminoglycoside antibiotic, induces apoptosis in the proximal tubule epithelium of rats treated at low, therapeutically relevant doses (El Mouedden et al., Antimicrob. Agents Chemother. 44, 665-675, 2000). Renal cell lines (LLC-PK(1) and MDCK-cells) have been used to further characterize and quantitate this process (electron microscopy; terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick-end labeling of fragmented DNA [TUNEL]; and DNA size analysis [oligonucleosomal laddering]). Cells were exposed for up to 4 days to gentamicin concentrations of up to 3 mM. Apoptosis developed, almost linearly, with time and drug concentration, and was (i) preventable within the time-frame of the experiments by overexpression of the anti-apoptotic protein Bcl-2, and by co-incubation with cycloheximide (MDKC but not LLC-PK(1) cells); (ii) associated with an increased activity of caspases (MDCK cells; bcl-2 transfectants showed no increase of caspase activities and Z-VAD.fmk afforded full protection). Gentamicin-induced apoptosis also developed to a similar extent in embryonic fibroblasts cultured under the same conditions. In the 3 cell types, apoptosis (measured after 4 days) was directly correlated with cell gentamicin content (apoptotic index [approximately 10 to 18% of TUNEL (+) cells for a content of 20 microg of gentamicin/mg protein; kidney cortex of rats showing apoptosis in proximal tubule epithelium typically contains approximately 10 microg of gentamicin/mg protein). Thus, gentamicin has an intrinsic capability of inducing apoptosis in eucaryotic cells. Development of apoptosis in proximal tubules of kidney cortex in vivo after gentamicin systemic administration is therefore probably related to its capacity to concentrate in this epithelium after systemic administration.

Apoptosis in renal proximal tubules of rats treated with low doses of aminoglycosides.

Kidney cortex apoptosis was studied with female Wistar rats treated for 10 days with gentamicin and netilmicin at daily doses of 10 or 20 mg/kg of body weight and amikacin or isepamicin at daily doses of 40 mg/kg. Apoptosis was detected and quantitated using cytological (methyl green-pyronine) and immunohistochemical (terminal deoxynucleotidyltransferase-mediated dUTP-biotin nick end labeling) staining, in parallel with a measurement of drug-induced phospholipidosis (cortical phospholipids and phospholipiduria), cortical proliferative response ((3)H incorporation in DNA and histoautoradiography after in vivo pulse-labeling with [(3)H]thymidine), and kidney dysfunction (blood urea nitrogen and creatinine). Gentamicin induced in proximal tubules a marked apoptotic reaction which (i) was detectable after 4 days of treatment but was most conspicuous after 10 days, (ii) was dose dependent, (iii) occurred in the absence of necrosis, and (iv) was nonlinearly correlated with the proliferative response (tubular and peritubular cells). Comparative studies revealed a parallelism among the extents of phospholipidosis, apoptosis, and proliferative response for three aminoglycosides (gentamicin >> amikacin congruent with isepamicin). By contrast, netilmicin induced a marked phospholipidosis but a moderate apoptosis and proliferative response. We conclude that rats treated with gentamicin develop an apoptotic process as part of the various cortical alterations induced by this antibiotic at low doses. Netilmicin, and still more amikacin and isepamicin, appears safer in this respect. Whereas a relation between aminoglycoside-induced tubular apoptosis and cortical proliferative response seems to be established, no simple correlation with phospholipidosis can be drawn.