Acta Physiol Scand 1990, 141, 27-34
Gentamicin inhibition of Na+,K+-ATPase in rat kidney cells Y. F U K U D A , A.-S. M A L M B O R G " and A. A P E R I A Departments of Pediatrics and * Clinical Bacteriology, Karolinska Institute, Stockholm, Sweden FUKUDA, Y., MALMBORG, A.-S. & APERIA,A. 1990. Gentamicin inhibition of Na+,K+ATPase in rat kidney cells. Acta Physiol Scand 141, 27-34. Received 26 January 1990, accepted 26 July 1990. ISSN 0001-6772. Departments of Pediatrics and Clinical Bacteriology, Karolinska Institute, Stockholm, Sweden. Na,K+-ATPase activity is decreased in homogenized renal tissue from GM-treated rats. This study examines whether the site of the active effect of GM on Na,K-ATPase activity in the kidney can be localized to the proximal convoluted tubules (PCT) where the drug is taken up and where it will produce necrosis. In rats treated with gentamicin (50 pg. kg-'.day-l i.m.) for 7 days, PCT Na,K-ATPase activity was reduced as compared to vehicle-treated rats but returned to control levels 7 days after treatment withdrawal. In another nephron segment, the medullary thick ascending limb of Henle (mTAL), where GM induced lesions are uncommon, Na,K-ATPase activity was the same in GM- and vehicle-treated rats treatment. T o study the in vitro effect of GM, dissected PCT and mTAL segments from untreated rats were preincubated for 30 min with GM M, a dose similar to the tissue concentration in chronically treated rats. In tubule segments that were permeabilized to allow the drug to enter the cells, GM M significantly inhibited Na,K-ATPase activity both in PCT and mTAL. In non-permeabilized mTAL segments GM did not inhibit Na,K-ATPase activity. GM inhibition of Na,K-ATPase activity in permeabilized PCT segments persisted after the tubules were rinsed in GM free medium. G M does not inhibit Na,K-ATPase partly purified from the renal cortex. Conclusion. Gentamicin inhibits Na,K-ATPase activity in renal tubule cells when it has access to the cytoplasm. Treatment with GM will therefore cause a selective inhibition of Na,K-ATPase in the proximal tubule cells. Key words : aminoglycosides, Na,K-ATPase, nephrotoxicity, proximal tubule, thick ascending limb
T h e nephrotoxicity of gentamicin ( G M ) and other aminoglycosides is well recognized (Kaloyanides & Pastoriza-Munoz 1980). Chronic GM treatment can cause necrosis and death of tubule cells (Kaloyanides & Pastoriza-Munoz 1980). T h e injury is almost exclusively confined to proximal tubule ( P T ) cells where the drug accumulates (Silverblatt & Kuehn 1979, Morin et al. 1980, Vandewalle et al. 1981, Wedeen et al. 1983). Correspondence : Anita Aperia MD, Department of Pediatrics, St Goran's Children's Hospital, S-112 81 Stockholm, Sweden.
I n renal cortical homogenates from rats chronically treated with GM, Na,K-ATPase activity is decreased (Williams et al. 1981, 1984, Cronin et al. 1982, Cronin & Newman 1985). I t has been suggested that changes in the activity of Na,K-ATPase might contribute to GM nephrotoxicity. T h e site of action for GM inhibition of Na,K-ATPase has, however, not been localized. T h i s study examines the effect of GM on Na,K-ATPase activity in proximal tubule segments and medullary thick ascending limb (mTAL) segments both in vivo and in vitro. I n m T A L GM uptake is negligible and the cells are considered to be resistant to GM toxicity.
27
28
Y . Fukudn et al.
W e found that GM only inhibited Na,K-ATPase in proximal convoluted tubule the in-vitro studies (PCT) segments in viva.
GM also inhibited N ~ , K - A T in ~the~ m~ ~~ ~ segments provided that they were made permeable prior to GM incubation. M A T E R I A L S AND METHODS Experimental animals and treatment protocol. Young adult (40-53 days) male SpragueDawley rats were used. The body weight of rats aged 40 days ranged from 165 to 200 g. Rat chow containing 21 yo protein (Ewos, Sodertalje, Sweden) and water was given ad libitum. In studies of the chronic effect of G M the rats were injected intramuscularly with G M 50 mg kg-' day-l (25 mg kg-l twice daily) for 7 days. The rats were studied at 1, 2 and 7 days after cessation of the therapy. T h e G M concentration in the renal cortex was determined on the last day of treatment in 10 rats. Those rats were not used for any other studies. The tissue was homogenized according to Zager & Prior (1986), and G M was determined with enzyme-mediated immunoassay (EMIT; White et al. 1981). The G M concentration averaged M). In a few 341 14 pg mg-' wet tissue (0.7 x rats, not used for any other studies, blood was withdrawn for determination of serum creatinine. Analysis of serum creatinine was made with a Beckman creatinine analyser. Preparation of tubules. In each study the rats were anaesthetized with an intraperitoneal injection of Inactin 80 mg kg-' (Byk-Gulden, Konstanz, FRG). Following a midline incision, the left kidney was exposed and perfused, as described previously (Rane & Aperia 1985), rinsed with Ringer solution and then with a modified Hank's solution (containing in mM: NaCl 137, KCI 5, MgSo, 0.8, NaHPO, 0.33, KH,PO, 0.44, CaCI, 1, MgCI, 1, Tris-HCI 10) to which collagenase 0.05% (Sigma, St Louis, MO, USA) and bovine serum albumin (BSA) 0.1% (Behringwerke AG, Marburg, FRG) were added. The pH was adjusted to 7.4. Kidney blood flow was not interrupted prior to the perfusion. The kidney was removed and thin pyramids were cut along the cortical papillary axis and incubated at 35 "C in a modified Hank's solution containing collagenase and bubbled with air. The incubation time varied between 5 and 20 min. The tissue was then transferred to a solution with the same composition as the perfusion solution except that collagenase and BSA were omitted and the CaCI, concentration was 0.25 mM. The PCT segments and mTAL segments were dissected from the superficial nephrons and the outer medulla respectively under a stereomicroscope at 4 "C. The tubule segments were individually transferred to the con-
cavity of a bacteriological slide, inspected and photographed for length determinations, using an inverted microscope at x 100 magnification. The tubules I , were then stored on ice for between 30 and 60 min. Incubation of tubules with gentamicin. The tubules were transferred to room temperature, permeabilized either with a brief hypotonic shock (PCT) or with saponin 0.01yo(mTAL) and preincubated for 30 min with or without M GM. mTAL segments that had not been permeabilized prior to preincubation were also studied. This protocol was not used for PCT segments since the membrane of these cells is less tight. PCT segments are somewhat permeable to [32P]ATP even before the hypotonic shock. T o determine whether the effect of G M on Na,K-ATPase activity was reversible, PCT segments M for 30 min that had been incubated in G M were rinsed several times in GM-free microdissection solution and then kept in a GM-free microdissection solution for another 30 min. Determination of Na,K-ATPase activity. The Na,K-ATPase activity was measured by a method previously described (Doucet et al. 1979, Katz et al. 1979, Rane & Aperia 1985). The tubule segments were made permeable with hypotonic shock, freezing and thawing, or saponin. They were incubated in a medium containing (in mM) NaCl SO, KCI 5, MgCI, 10, EGTA 1, Tris-HCI 100, Na, ATP (grade 11, Sigma) 10 and 13'P]ATP 5 pCi ml-' (New England Nuclear, 2-10 Ci mmol-'). For determination of ouabain-insensitive (Mg-dependent) ATPase activity, NaCl and KCI were omitted, Tris-HCI was 150 mM and 1 mM ouabain (Merck, Darmstadt, FRG) was added. The pH of both solutions was 7.4. Media containing the tubule segments and the medium alone (blank) were incubated at 37 "C for 15 min. After absorption of the unhydrolysable nucleotide on activated charcoal, the 32Pliberated by the hydrolysis of [32P]ATP was separated by filtration through a millipore filter (HA, 0.45 pm, France). The radioactivity was counted in a liquid scintillation spectrophotometer. In each study, we determined total ATPase and ouabain-insensitive ATPase in 6-10 segments. Four to six determinations of 32P released from [32P]ATPin incubation solutions containing no tubule segments were made to correct for non-specific ATP hydrolysis in each assay. Non-specific ATP hydrolysis was consistently less than 10%. The mean value of these determinations (blanks) was subtracted from each value for 32P released from tubular incubations. In each experiment Na,K-ATPase activity was calculated as the difference between the mean value for total ATPase and the mean value for ouabain-insensitive activity ATPase and expressed per unit of tubule length in a single tubule. Effect of GM M on partly purtjied Na,K-ATPase. Na,K-ATPase was purified from
3000
2000
I
-NS-
3000
20001000 IOOO-
0
0.. Day after therapy withdrawal
I 2 7 Day after therapy withdrawal
Fig. 1. In-vivu effect of GM treatment (25 mg kg-' twice daily for 7 days) on Na,K-ATPase activity (pmol Pj mm-' tubule h-' in PCT (a) and mTAL (b) from adult rats. Na,K-ATPase activity in tubule segments from GM-treated rats (hatched bars) and vehicle-treated rats (unfilled bars) was determined at 1, 2 and 7 days after treatment withdrawal. Values are meansfSE; +*P < 0.01, ***P < 0,001, NS, not significant compared with vehicle treatment; n = 4-5 rats in each group.
the renal cortex of adult Sprague-Dawley rats using the procedure described by Jergensen (1974). For determination of enzyme activity 10-pl aliquots of purified enzyme were incubated for 15 min in 90 pl of medium containing final concentrations of 5 mM KCI, 5 mM MgCI,, 1 mM EGTA, 30 mM Tris-HCI, 3 mM ATP and a tracer amount of [T-~'P]ATPin the presence or absence of G M 1 0 - 3 ~with various concentrations of sodium (from 0 to 130 mM) at 37 "C (pH 7.4). The amount of enzyme was selected precisely in order to hydrolyse less than 20% of total ATP. The reaction was stopped by the addition of 700 pl of a cold-activated charcoal, followed by centrifugation. The phosphate liberated by hydrolysis of [T-~'P]ATP was determined in the supernatant. For determination of ouabain-insensitive ATPase activity, NaCl and KCI were omitted, and 1 mM ouabain was added. Na,K-ATPase was calculated as the difference between total ATPase and ouabaininsensitive ATPase activity. All data were subjected to non-linear least-squares analysis. Na+ dependence of the Na,K-ATPase activity was fitted by the cooperative model described x [Na+]")/(K+ "a+]"). by Lytton (1985) : v = (Pax VmaxrK and n (the Hill coefficient) were allowed to vary to obtain the best fit for the data. The apparent for Na was obtained from calculated values of K and n with the following relationship: apparent
= (K)''=. Protein concentration was measured by the method of Lowry et al. (1951). Calculations. Values are given as means fSE. The Student t-test and analysis of variance were used for comparisons between the groups. P-values < 0.05 were considered significant.
RESULTS Chronic effects of GM on Na,K-ATPase activity in renal tubules
T h e rats that were treated for 7 days with GM remained in good health throughout the treatment as judged from appetite, general behaviour and reaction to anaesthesia. Serum creatinine was generally higher in GM-treated rats than in vehicle-treated rats. Body weight was the same or slightly lower in GM-treated rats than in vehicle-treated rats 2 days after cessation of therapy compared with vehicle-treated rats ( P
Gentamicin inhibition of Na+,K+-ATPase in rat kidney cells Y. F U K U D A , A.-S. M A L M B O R G " and A. A P E R I A Departments of Pediatrics and * Clinical Bacteriology, Karolinska Institute, Stockholm, Sweden FUKUDA, Y., MALMBORG, A.-S. & APERIA,A. 1990. Gentamicin inhibition of Na+,K+ATPase in rat kidney cells. Acta Physiol Scand 141, 27-34. Received 26 January 1990, accepted 26 July 1990. ISSN 0001-6772. Departments of Pediatrics and Clinical Bacteriology, Karolinska Institute, Stockholm, Sweden. Na,K+-ATPase activity is decreased in homogenized renal tissue from GM-treated rats. This study examines whether the site of the active effect of GM on Na,K-ATPase activity in the kidney can be localized to the proximal convoluted tubules (PCT) where the drug is taken up and where it will produce necrosis. In rats treated with gentamicin (50 pg. kg-'.day-l i.m.) for 7 days, PCT Na,K-ATPase activity was reduced as compared to vehicle-treated rats but returned to control levels 7 days after treatment withdrawal. In another nephron segment, the medullary thick ascending limb of Henle (mTAL), where GM induced lesions are uncommon, Na,K-ATPase activity was the same in GM- and vehicle-treated rats treatment. T o study the in vitro effect of GM, dissected PCT and mTAL segments from untreated rats were preincubated for 30 min with GM M, a dose similar to the tissue concentration in chronically treated rats. In tubule segments that were permeabilized to allow the drug to enter the cells, GM M significantly inhibited Na,K-ATPase activity both in PCT and mTAL. In non-permeabilized mTAL segments GM did not inhibit Na,K-ATPase activity. GM inhibition of Na,K-ATPase activity in permeabilized PCT segments persisted after the tubules were rinsed in GM free medium. G M does not inhibit Na,K-ATPase partly purified from the renal cortex. Conclusion. Gentamicin inhibits Na,K-ATPase activity in renal tubule cells when it has access to the cytoplasm. Treatment with GM will therefore cause a selective inhibition of Na,K-ATPase in the proximal tubule cells. Key words : aminoglycosides, Na,K-ATPase, nephrotoxicity, proximal tubule, thick ascending limb
T h e nephrotoxicity of gentamicin ( G M ) and other aminoglycosides is well recognized (Kaloyanides & Pastoriza-Munoz 1980). Chronic GM treatment can cause necrosis and death of tubule cells (Kaloyanides & Pastoriza-Munoz 1980). T h e injury is almost exclusively confined to proximal tubule ( P T ) cells where the drug accumulates (Silverblatt & Kuehn 1979, Morin et al. 1980, Vandewalle et al. 1981, Wedeen et al. 1983). Correspondence : Anita Aperia MD, Department of Pediatrics, St Goran's Children's Hospital, S-112 81 Stockholm, Sweden.
I n renal cortical homogenates from rats chronically treated with GM, Na,K-ATPase activity is decreased (Williams et al. 1981, 1984, Cronin et al. 1982, Cronin & Newman 1985). I t has been suggested that changes in the activity of Na,K-ATPase might contribute to GM nephrotoxicity. T h e site of action for GM inhibition of Na,K-ATPase has, however, not been localized. T h i s study examines the effect of GM on Na,K-ATPase activity in proximal tubule segments and medullary thick ascending limb (mTAL) segments both in vivo and in vitro. I n m T A L GM uptake is negligible and the cells are considered to be resistant to GM toxicity.
27
28
Y . Fukudn et al.
W e found that GM only inhibited Na,K-ATPase in proximal convoluted tubule the in-vitro studies (PCT) segments in viva.
GM also inhibited N ~ , K - A T in ~the~ m~ ~~ ~ segments provided that they were made permeable prior to GM incubation. M A T E R I A L S AND METHODS Experimental animals and treatment protocol. Young adult (40-53 days) male SpragueDawley rats were used. The body weight of rats aged 40 days ranged from 165 to 200 g. Rat chow containing 21 yo protein (Ewos, Sodertalje, Sweden) and water was given ad libitum. In studies of the chronic effect of G M the rats were injected intramuscularly with G M 50 mg kg-' day-l (25 mg kg-l twice daily) for 7 days. The rats were studied at 1, 2 and 7 days after cessation of the therapy. T h e G M concentration in the renal cortex was determined on the last day of treatment in 10 rats. Those rats were not used for any other studies. The tissue was homogenized according to Zager & Prior (1986), and G M was determined with enzyme-mediated immunoassay (EMIT; White et al. 1981). The G M concentration averaged M). In a few 341 14 pg mg-' wet tissue (0.7 x rats, not used for any other studies, blood was withdrawn for determination of serum creatinine. Analysis of serum creatinine was made with a Beckman creatinine analyser. Preparation of tubules. In each study the rats were anaesthetized with an intraperitoneal injection of Inactin 80 mg kg-' (Byk-Gulden, Konstanz, FRG). Following a midline incision, the left kidney was exposed and perfused, as described previously (Rane & Aperia 1985), rinsed with Ringer solution and then with a modified Hank's solution (containing in mM: NaCl 137, KCI 5, MgSo, 0.8, NaHPO, 0.33, KH,PO, 0.44, CaCI, 1, MgCI, 1, Tris-HCI 10) to which collagenase 0.05% (Sigma, St Louis, MO, USA) and bovine serum albumin (BSA) 0.1% (Behringwerke AG, Marburg, FRG) were added. The pH was adjusted to 7.4. Kidney blood flow was not interrupted prior to the perfusion. The kidney was removed and thin pyramids were cut along the cortical papillary axis and incubated at 35 "C in a modified Hank's solution containing collagenase and bubbled with air. The incubation time varied between 5 and 20 min. The tissue was then transferred to a solution with the same composition as the perfusion solution except that collagenase and BSA were omitted and the CaCI, concentration was 0.25 mM. The PCT segments and mTAL segments were dissected from the superficial nephrons and the outer medulla respectively under a stereomicroscope at 4 "C. The tubule segments were individually transferred to the con-
cavity of a bacteriological slide, inspected and photographed for length determinations, using an inverted microscope at x 100 magnification. The tubules I , were then stored on ice for between 30 and 60 min. Incubation of tubules with gentamicin. The tubules were transferred to room temperature, permeabilized either with a brief hypotonic shock (PCT) or with saponin 0.01yo(mTAL) and preincubated for 30 min with or without M GM. mTAL segments that had not been permeabilized prior to preincubation were also studied. This protocol was not used for PCT segments since the membrane of these cells is less tight. PCT segments are somewhat permeable to [32P]ATP even before the hypotonic shock. T o determine whether the effect of G M on Na,K-ATPase activity was reversible, PCT segments M for 30 min that had been incubated in G M were rinsed several times in GM-free microdissection solution and then kept in a GM-free microdissection solution for another 30 min. Determination of Na,K-ATPase activity. The Na,K-ATPase activity was measured by a method previously described (Doucet et al. 1979, Katz et al. 1979, Rane & Aperia 1985). The tubule segments were made permeable with hypotonic shock, freezing and thawing, or saponin. They were incubated in a medium containing (in mM) NaCl SO, KCI 5, MgCI, 10, EGTA 1, Tris-HCI 100, Na, ATP (grade 11, Sigma) 10 and 13'P]ATP 5 pCi ml-' (New England Nuclear, 2-10 Ci mmol-'). For determination of ouabain-insensitive (Mg-dependent) ATPase activity, NaCl and KCI were omitted, Tris-HCI was 150 mM and 1 mM ouabain (Merck, Darmstadt, FRG) was added. The pH of both solutions was 7.4. Media containing the tubule segments and the medium alone (blank) were incubated at 37 "C for 15 min. After absorption of the unhydrolysable nucleotide on activated charcoal, the 32Pliberated by the hydrolysis of [32P]ATP was separated by filtration through a millipore filter (HA, 0.45 pm, France). The radioactivity was counted in a liquid scintillation spectrophotometer. In each study, we determined total ATPase and ouabain-insensitive ATPase in 6-10 segments. Four to six determinations of 32P released from [32P]ATPin incubation solutions containing no tubule segments were made to correct for non-specific ATP hydrolysis in each assay. Non-specific ATP hydrolysis was consistently less than 10%. The mean value of these determinations (blanks) was subtracted from each value for 32P released from tubular incubations. In each experiment Na,K-ATPase activity was calculated as the difference between the mean value for total ATPase and the mean value for ouabain-insensitive activity ATPase and expressed per unit of tubule length in a single tubule. Effect of GM M on partly purtjied Na,K-ATPase. Na,K-ATPase was purified from
3000
2000
I
-NS-
3000
20001000 IOOO-
0
0.. Day after therapy withdrawal
I 2 7 Day after therapy withdrawal
Fig. 1. In-vivu effect of GM treatment (25 mg kg-' twice daily for 7 days) on Na,K-ATPase activity (pmol Pj mm-' tubule h-' in PCT (a) and mTAL (b) from adult rats. Na,K-ATPase activity in tubule segments from GM-treated rats (hatched bars) and vehicle-treated rats (unfilled bars) was determined at 1, 2 and 7 days after treatment withdrawal. Values are meansfSE; +*P < 0.01, ***P < 0,001, NS, not significant compared with vehicle treatment; n = 4-5 rats in each group.
the renal cortex of adult Sprague-Dawley rats using the procedure described by Jergensen (1974). For determination of enzyme activity 10-pl aliquots of purified enzyme were incubated for 15 min in 90 pl of medium containing final concentrations of 5 mM KCI, 5 mM MgCI,, 1 mM EGTA, 30 mM Tris-HCI, 3 mM ATP and a tracer amount of [T-~'P]ATPin the presence or absence of G M 1 0 - 3 ~with various concentrations of sodium (from 0 to 130 mM) at 37 "C (pH 7.4). The amount of enzyme was selected precisely in order to hydrolyse less than 20% of total ATP. The reaction was stopped by the addition of 700 pl of a cold-activated charcoal, followed by centrifugation. The phosphate liberated by hydrolysis of [T-~'P]ATP was determined in the supernatant. For determination of ouabain-insensitive ATPase activity, NaCl and KCI were omitted, and 1 mM ouabain was added. Na,K-ATPase was calculated as the difference between total ATPase and ouabaininsensitive ATPase activity. All data were subjected to non-linear least-squares analysis. Na+ dependence of the Na,K-ATPase activity was fitted by the cooperative model described x [Na+]")/(K+ "a+]"). by Lytton (1985) : v = (Pax VmaxrK and n (the Hill coefficient) were allowed to vary to obtain the best fit for the data. The apparent for Na was obtained from calculated values of K and n with the following relationship: apparent
= (K)''=. Protein concentration was measured by the method of Lowry et al. (1951). Calculations. Values are given as means fSE. The Student t-test and analysis of variance were used for comparisons between the groups. P-values < 0.05 were considered significant.
RESULTS Chronic effects of GM on Na,K-ATPase activity in renal tubules
T h e rats that were treated for 7 days with GM remained in good health throughout the treatment as judged from appetite, general behaviour and reaction to anaesthesia. Serum creatinine was generally higher in GM-treated rats than in vehicle-treated rats. Body weight was the same or slightly lower in GM-treated rats than in vehicle-treated rats 2 days after cessation of therapy compared with vehicle-treated rats ( P
