Pediatr Nephrol (1994) 8: 447- 450 9 IPNA 1994

Pediatric Nephrology

Original article
Transplacental effects of gentamicin on endocytosis in rat renal proximal tubule cells
Hatem Smaouil, Madeleine Schaeverbeke 1, Jean-Pierre Malli6 2, and Jean Schaeverbekel
1Laboratoire de Biologic Cellulaire, Universitd Paris 7, Tour 23-33, 1er 6t, 2 place Jussieu, F-75251 Paris Cedex 05, France 2 Laboratoire de N6phrologie, Universit6 de Nancy I, BP 184, F-54505 Vandoeuvre Cedex, France Received July 23, 1993; received in revised form December 9, 1993; accepted January 7, 1994

Abstract. Changes in kidney maturation in utero have been

reported after gentamicin administratio~ to pregnant rats. While the proteinuria commonly observed could be related to modifications of the glomerular basement membrane, perturbed renal protein handling could be accounted for by changes in the proximal tubular cells. Therefore, we studied the effect of gentamicin on the renal handling and transport of proteins in proximal tubular ceils using the horseradish peroxidase, a fluid-phase marker, as a probe. Gentamicin was administered intraperitoneally to pregnant Wistar rats (75 mg/kg body weight per day) and neonatal kidneys were studied 1 day after birth. In proximal tubular cells of the deep cortical area, containing the fully matured nephrons of neonates, the transport and digestion of reabsorbed peroxidase was considerably reduced compared with controls where peroxidase reached lysosomes after endocytosis. Urinary protein excretion increased in treated animals. We conclude that gentamicin, entering the proximal tubular cells via the endocytic pathway, decreases the tubular reabsorption of proteins, thus increasing urinary protein excretion.

Key words: Gentamicin - Proximal tubular cells - Endocytosis - Gestation - Neonate

Introduction
Aminoglycoside antibiotics are widely used in the treatment of gram-negative infections, but are often associated with the development of acute renal failure [1-4]. Gentamicin is excreted by glomerular filtration and partly reabsorbed by pinocytosis in the proximal tubular cells (PTC) [5-8], where it accumulates in lysosomes. The first morphological change is the formation of myeloid bodies

which consist of osmiophilic material arranged in multilamellar layers. These myeloid bodies represent an impairment of the biodegradation of complex polar lipids [9-11]. Depression of the glomerular filtration rate, proteinuria and enzymuria are other manifestations of this nephrotoxicity [ 12 - 14]. Aminoglycoside nephrotoxicity has been largely described in adult humans and experimental animals [15-17]. In the last decade, effects of aminoglycosides on kidney development have been reported when females were given these antibiotics during pregnancy [ 18-22]. Proteinuria is a common feature of aminoglycoside-induced nephrotoxicity [15]; in adults it was attributed mainly to tubular cell desquamation and a change in glomerular permselectivity. In neonates exposed to aminoglycosides in utero, we previously reported that the proteinuria was due to an altered metabolism of the glomerular basement membrane (GBM) [22, 23]. Since gentamicin is given to pregnant women [24, 25] and is able to cross the placenta [26], we have studied its effects on the development of rat PTC. While investigations in both adults and children have pointed out disturbances in renal structure, tubular cell function has not been studied. Therefore, we focused on the intracellular distribution of gentamicin and its effects on protein handling. In order to investigate whether gentamicin kvas absorbed by PTC when the antibiotic was given during pregnancy we applied a specific anti-gentamicin antibody to thin sections of neonate kidneys and used protein A-gold as a marker. Reabsorption of proteins in neonatal kidneys was investigated using horseradish peroxidase (HRP), intravenously injected, as a probe.

Materials and methods

Animals and treatments.

Correspondence to:

H. Smaoui

Female Wistar rats were mated overnight; the day on which sperm was found in the vaginal smear was considered as day 0 of gestation. Animals were then individually housed in plastic cages and fed a standard diet (UAR 113). Water was allowed ad libitum. Females were given gentamicin (Gentalline, Laboratoire Uni-

448 cet, Paris) daily from day 7 to 11 (organogenesis) and from day 14 to 18 (beginning of nephrogenesis) of gestation, or saline by i.p. injection at 4 p.m. Twelve mothers received gentamicin (75 mg/kg body weight per day), 120 neonates; saline was administered to 10 pregnant control rats (100 neonates). Delivery occurred normally on day 21 of gestation. Neonates were studied on the following day. Four neonates per litter were used for experiments with HRP and 4 others were used for immunocytochemistry.

Immunocytochemistry. Kidney slices were fixed by immersion in 4%

paraformaldehyde and 0.5% glutaraldehyde in 0.12 M cacodylate buffer plus 0.043 M sodium chloride (NaC1), pH 7.40; they were then rinsed in buffel; dehydrated in alcohol at -20 ~ C and embedded in the hydrophilic resin, LRWhite, at -20 ~ C. Sections were cut with a diamond knife and picked up on Parlodion-coated nickel grids. Sections were first incubated for 10 rain on a drop of 0.1% gelatin in phosphate-buffered saline (PBS, 0.9% NaC1 in 10 mM sodium phosphate buffer, pH 7.40) to reduce non-specific binding of antibodies. The grids were then washed in the same buffer and transferred to a drop of polyclonal rabbit anti-gentamicin antibody (ICN) diluted 1:50 and incubated for 1 h at room temperature in a moist chamber. After three 5-min rinses in PBS, the grids were incubated for 1 h on a drop of protein A-gold solution containing 5-nm gold particles prepared by density gradient ceutrifugation according to Slot and Gueuze [27]. Control grids were stained by omitting the first incubation step with the antibody. Grids were then washed with PBS, contrasted with uranyl acetate and observed with a Philips CM12 electron microscope.

Fig. 1. Localization of gentamicin with anti-gentamicin antibody and protein A-gold (5 nm) in a glomemlus from a neonate exposed to gentamicin in utero. Labelling is seen in capillary (CAP),in glomerular basement membrane (GBM) and in urinary space (US). Arrows, gold particles, x50,000

Experiments with HRP. The protein tracer HRP (type II, Sigma, molecular weight 40,000) was dissolved in saline and perfused via the jugular vein in a small volume to prevent haemodynamic changes. Thus neonatal rats received 0.2 mg HRP/g body weight and were killed 15 min after injection of the tracer. Kidneys were removed and cut into slices 4-5 mm thick, which were then fixed for 3 h in 4% paraformaldehyde, 0.5% glutaraldehyde in 0.12 M sodium cacodylate buffer plus 0.043 NaC1, pH 7.40. After washing three times in 0.12 M cacodylate buffer slices were immersed in the same medium plus dimethyl sulphoxide (10%) for 1 h at 4~ C, then frozen in a solution of isopentane in liquid nitrogen. Sections (20 gm) were cut with a cryostat microtome (American Optical), washed in 0.1 M TRIS-HC1pH 7.40 and incubated for 10 min in a filtered solution of 0.01 M 3-3' diaminobenzidine in 0.01 M TRIS-HCI buffer, pH 7.40 containing 0.01% hydrogen peroxide. The sections were then washed in three changes of the same buffer and post-fixed for 30 miu in 2% ferrocyanide-reduced osmium in cacodylate buffer; they were dehydrated in graded alcohols, in propylene oxide and embedded in Epon 812. The sections were cut with a Reicbert-Jung ultramicrotome, stained with uranyl acetate and lead and examined with a Philips CM12 electron microscope.

len and, next to these areas, membranes appeared closely packed, resembling a small myeloid body.

Labelling with the anti-gentamicin antibody

The anti-gentamicin antibody was localized with protein A-gold; electron-dense gold particles were visualized by electron microscopy. Gentamicin was located in the glomerular capillaries, in the G B M and urinary spaces of treated animals (Fig. 1). In PTC, some scattered clusters of gold particles were found in the brush border and in compartments at the periphery of the cell membrane and in the perinuclear zone (Fig. 2b). No gold particles were seen in controls (Fig. 2 a).

Localization of peroxidase in PTC

In controls, HRP was taken into small vesicles, transferred into early endosomes and late endosomes and accumulated exclusively in lysosomes. Differences in the intensity of staining of the vesicles indicated differences in HRP concentrations in these structures (Fig. 3 a). In treated animals, the brush border membranes showed an increased electron opacity, representing the adsorption of peroxidase. In addition, many of the apical tubular invaginations at the base of the brush border microvilli contained dense accumulations of reaction product, while early endosomes did not contain peroxidase (Fig. 3 b).

Measurement of urinary proteins. Urine samples from neonates were

obtained by bladder puncture, and urinary proteins were measured by the procedure of Lowry et al. [28] using bovine serum albumin as a standard.

Statistical analysis. Differences between means were assessed by

Student's t-test.

Results

Proximal tubular cell morphology Urinalysis

Only the deep cortical area, containing the fully differentiated neonatal nephrons, was examined. Subcellular lesions were found in epithelial cells of animals exposed to gentamicin in utero: lysosomes frequently contained myeloid bodies, Golgi apparatus cisternae were locally swolA n increase in urinary proteins was observed in gentamicin-treated rats compared with controls. Mean values were 11.10 __ 0.040 and 6.40 + 0,012 gg/g body weight, respectively (P < 0.05).

449

Fig. 2. Localization of gentamicin with anti-gentamicin antibody and protein A-gold (5 nm) in proximal tubular cells (PTC) from a control and b gentamicin-treated animals. There is no labelling in controls; in treated animals labelling is present in brush border (BB) and in endosomes (E). Arrows, gold particles, x50,000

Fig. 3. Horseradish peroxidase (HRP) localization in the fully matured PTC from a control and b animals exposed to gentamicin in utero. In controls HRP is present in BB, in small vesicles, in early endosomes (EE) and late endosomes (LE). In treated animals HRP is only present in BB and in some EE. These alterations are not focal and are seen in all observed PTC (four neonates examined per litter). Arrowhead, myeloid bodies, x15,000

Discussion
It is well known that the rat kidney is not completely matured at birth, so that all stages of nephron differentiation can be observed in the same cortex from capsular to juxtamedullary regions. We studied the fully differentiated nephrons which, at this age, were located only in the deep cortex. In neonates born to treated mothers, the gentamicin filtered through the GBM, since anti-gentamicin antibody was localized in glomerular capillaries, in G B M and in urinary spaces. In PTC the labelling pattern was similar to

that previously observed in adults [6, 7, 29]: the gentamicin was located in brush border and in endocytic vacuoles. This study also indicated that gentamicin led to a defect of protein reabsorption by the PTC and inhibited the passage from apical invaginations to early and late endosomes, and consequently to lysosomes. HRP was only seen in brush border and in some apical vesicles of the very first step of endocytosis. The decreased tubular uptake of HRP in gentamicin-treated animals could be interpreted as a decreased rate of endocytosis, and the accumulation of HRP in the brush border of PTC may be secondary to an inhibition of the delivery of proteins to early endosomes.

450 As indicated by tracer studies, PTC from the gentamicin group were unable to incorporate HRP into endocytic vesicles. At the same age, the endocytic process occurred normally in the control fully differentiated PTC. H R P was absorbed by the endocytic vacuoles and never accumulated in the cytoplasm; the reaction product was seen in brush border, in coated pits and vesicles, in early and late endosomes and in lysosomes. Therefore, gentamicin treatment in utero completely or partially inhibited the endocytic process. In normal rat fetuses, protein reabsorption in the fully differentiated PTC begins at day 18 o f gestation (which lasts 21 days) [30]. Our study showed that endocytosis in 1-day-old neonates exposed to gentamicin in utero was similar to that in normal fetuses 3 days before birth. Therefore gentamicin is likely to slow down the maturation of endocytic protein reabsorption. We have already reported that in neonates exposed to gentamicin in utero the increased proteinuria is largely due to a delay in the maturity of the glomerular filtration barrier [22, 23]. We demonstrate in this study that this proteinuria may also be related to an immaturity of the tubular reabsorption system. In conclusion, gentamicin given to the mother during pregnancy leads, in neonates, to a decrease of the PTC endocytic process, which evokes a maturation delay and is partially responsible for the proteinuria. 9. Feldman S, Wang MY, Kaloyanides GJ (1982) Aminoglycosides induce a phospholipidosis in the renal cortex of the rat: an early manifestation of nephrotoxicity. J Pharmacol Exp Ther 220: 514-520 10. Hosteter KY, Hall LB (1989) Inhibition of kidney lysosomal phospholipase A and C by aminoglycoside antibiotic: possible mechanism of aminoglycoside toxicity. Proc Natl Acad Sci USA 79:1663 - 1663 11. Toubeau G, Maldague R Laurent G, Vaamonde CA, Tulkens PM, Hneson Stiennon JA (1986) Morphological alterations in distal and collecting tubules of the rat renal cortex after aminoglycoside administration at low doses. Virchows Arch [B] 51:475-485 12. Bayliss C, Rennke HG, Brenner BM (1977) Mechanisms of the defects in glomerular ultrafiltration associated with gentamicin administration. Kidney Int 12:344-353 13. Welwood JM, Lovell D, Thompson AE, Tighe JR (1976) Renal damage caused by gentamicin: a study of the effects on renal morphology and urinary enzyme excretion. J Pathol 118:171 - 182 14. Schentag JJ, Sutfin TA, Plant ME, Jusko WJ (1978) Early detection of aminoglycoside nephrotoxicity with urinary [3~ microglobulin. J Med 9:201-210 15. Bennet WM (1983) Aminoglycoside nephrotoxicity. Nephron 35: 73 -77 16. Humes HD (1988) Aminoglycoside nephrotoxicity. Kidney Int 33: 900-911 17. Kaloyanides GJ, Pastoria-Munoz E (1980) AmJnoglycoside nephrotoxicity. Kidney Int 18:571-582 18. Gilbert T, Nabarra B, Merlet-Benichou C (1988) Light and electron-microscopic analysis of the kidney in newborn rats exposed to gentamicin in utero. Am J Pathol 130:33-43 19. Malli6 JR Coulon G, Billerey C, Faucourt A, Morin JP (1988) In utero aminoglycosides induced nephrotoxicity in rat neonates. Kidney Int 33:36-44 20. Malli6 JR Gerard H, Gerard A (1984) Gentamicin administration to pregnant rats: effect on fetal renal development in utero. Dev Pharmacol Ther 7 [Suppl 1]: 89-92 21. Malli6 JR Gerard H, Gerard A (1986) In-utero gentamicin-induced nephrotoxicity in rats. Pediatr Pharmacol 5:229-239 22. Smaoui H, Malli6 JR Cheignon M, Borot C, Schaeverbeke J (1991) Glomerular alterations in rat neonate after transplacental exposure to gentamicin. Nephron 59:626-631 23. Smaoui H, Malli6 JR Schaeverbeke M, Robert A, Schaeverbeke J (1993) Gentamicin administered during gestation alters glomemlar basement membrane development. Antimicrob Agents Chemother 37:1510-1517 24. Tessin I, Bergmack J, Hiesche K, Jagenburg R, Trollfors B (1982) Renal functions of neonates during gentamicin treatment. Arch Dis Child 57:758-760 25. Cowan RH, Jukkola AF, Arant BS Jr (1980) Pathophysiologic evidence of gentamicin nephrotoxicity in neonatal puppies. Pediatr Res 14:1204-1211 26. Yoshioka H, Monna T, Matsuda (1972) The placental transfer of gentamicin. J Pediatr 80: 121-123 27. Slot JW, Gueuze HJ (1981) Sizing of protein A-colloidal probes for immunoelectron microscopy. J Cell Biol 90:533-541 28. Lowry OH, Rosebrough NJ, FatT AL, Randall RJ (1951) Protein measurement with the folin phenol reagent. J Biol Chem 193: 265 -275 29. Wedeen RR Batuman V, Cheeks C, Marquet E, Sobel H (1983) Transport of gentamicin in rat proximal tubule. Lab Invest 48: 212-223 30. Schaeverbeke J, Cheignon M (1980) Differentiation of glomerular filter and tubular reabsorption apparatus during fetal development of the rat kidney. J Embryol Exp Morphol 58: 157-175

Acknowledgements. H. Smaoui is grateful to CNRS France for a PhD

fellowship.

References
1. Palco FG, Smith MM, Anieri GM (1969) Nephrotoxicity of aminoglycoside and gentamicin. J Infect Dis 119:406-409 2. De Broe ME, Paulus GJ, Verpooten GA, Roels F, Buyssens N, Wedden R, Van Hoof F, Tulkens PM (1984) Early effects of gentamicin, tobramycin and amikacin on the human kidney. Kidney Int 25:643-652 3. Provoost AR Adejuyighe O, Wolf ED (1985) Nephrotoxicity of aminoglycosides in young and adult rats. Pediatr Res 19: 1191-1196 4. Houghton DC, Hartnett M, Campbell-Boswell M, Porter G, Bennet WM (1976) A light and electron microscopic analysis of gentamicin nephrotoxicity in rats. Am J Pathol 82:589-612 5. Collier VU, Lietman PS, Mitch WE (1979) Evidence for luminal uptake of gemamicin in the perfused rat kidney. J Pharmacol Exp Ther 210:247-251 6. Silverblatt FJ, Kuehn C (1979) Autoradiography of gentamicin uptake by the rat proximal tubule cell. Kidney Iut 15:335-345 7. Sastrasinh M, Knanss TC, Weinberg JM, Humes D (1982) Identification of the aminoglycoside binding site in rat renal brush border membranes. J Pharmacol Exp Tber 222:350-358 8. Vandevalle A, Farman N, Morin JR Fillastre JR Hat PY, Bonvalet JP (1981) Gentamicin incorporation along the nephron: autoradiographic study on isolated tubules. Kidney Int 19:529-539