- Email: [email protected]
Evaluation of enzymuria as an indicator of amikacin-induced renal damage in guinea pigs
Toxicology Letters, 62 (1992) 101-114 6 1992 Elsevier Science Publishers B.V. All rights reserved 0378-4274/92/$5.00
101
TOXLET 02761
Evaluation of enzymuria as an indicator of amikacininduced renal damage in guinea pigs
Catherine A.M. Suzuki’, Barry H. Thomas and Ruedi Mueller’ Drug Toxicology Division, Bureau of Drug Research, Health Protection Branch, Sir F.G. Banting Research Centre, Tunney’s Pasture, Ottawa, Ontario (Canada)
(Received 23 January 1992) (Accepted 3 April 1992) Key words: Amikacin; Enzymuria; Nephrotoxicity
SUMMARY Guinea pigs were injected subcutaneously for 10 days with amikacin (AK) at a dose of 0, 100,200 or 400 mg/kg body wt. per day. The total daily dose was administered in either a single injection or divided equally and given as two daily injections. After the 10 days of AK treatment, uptake of the organic cation, tetraethylammonium (TEA) into renal cortical slices was inhibited in a dose-related manner. Changes in renal tubular morphology also increased with higher doses. The urinary excretion of the enzymes, N-acetyl/?-o-glucosaminidase (NAG), and alkaline phosphatase (ALP) significantly increased during the course of AK treatment, however, due to the large intragroup variability, the daily fluctuations and the absence of any distinct trends in urinary enzyme excretion it was difficult to establish a dose-relationship between AK-induced renal damage and the resultant enzymuria. At doses of 100 and 200 mg/kg body wt., the two-injection regimen resulted in the greater renal accumulation of AK and damage as reflected by a greater inhibition of TEA uptake and greater changes in renal tubular morphology. In contrast, this difference in toxicity could not be detected with enzymuria again due to the large intragroup variability and the absence of discernable excretion patterns of NAG and ALP. Thus, neither NAG nor ALP appear to be suitable quantitative markers of AK-induced nephrotoxicity.
‘Present address: Toxicology Research Division, Bureau of Chemical Safety, Health Protection Branch, Sir F.G. Banting Research Centre, Tunney’s Pasture, Ottawa, Ontario, Canada KlA 0L2. Correspondence to: C.A.M. Suzuki, Toxicology Research Division, Bureau of Chemical Safety, Health Protection Branch, Sir F.G. Banting Research Centre, Tunney’s Pasture, Ottawa, Ontario, Canada KlA OL2.
102
INTRODUCTION
The aminoglycoside (AG) antibiotics represent the first line of defence against a number of Gram-negative bacterial infections. Unfortunately, use of AGs may be limited due to a number of potential side effects including ototoxicity and nephrotoxicity [l]. Although the renal damage caused by these drugs is generally reversible, impairment of renal function during the course of therapy may, however, result in the reduction of AG clearance since 80-90% of these drugs are eliminated from the body via the kidney [l]. Reduced clearance may, in turn, result in sustained, elevated plasma drug concentrations [2], resulting in the accumulation of the drug to toxic levels in other sensitive target organs such as the ear. Consequently, monitoring the renal status of a patient during the course of AG treatment is of great importance. Currently, serum creatinine and/or blood urea nitrogen levels are used to monitor the renal function of patients undergoing AG therapy. However, these parameters are relatively insensitive markers of renal damage and become significantly altered only after considerable damage to the kidney has occurred [3]. Recently the measurement of enzymuria (renal-derived enzymes in the urine) has received some attention as a potential indicator of AG-induced nephrotoxicity [3-61. Although normally present only in low concentrations in the urine, following damage to the kidney, concentrations of certain enzymes in the urine can significantly increase. In some cases, their appearance in the urine may precede changes in traditional renal function tests [6]. Although the occurrence of enzymuria during gentamicin treatment has been reported with both experimental animals [7-lo] and patients receiving AGs [l l-141, the quantitative relationship between enzymuria and renal damage has not yet been well defined. Thus, the present study was undertaken to further examine this relationship using the AG, amikacin (AK). Amikacin-induced renal damage was assessed by measuring the urinary excretions of N-acetyl-/?-o-glucosaminidase (NAG) and alkaline phosphatase (ALP). To evaluate the usefulness of enzymuria as a quantitative marker of renal damage, organic ion transport in renal cortical slices was also measured due to the sensitivity of ion transport to AG-induced nephrotoxicity [15,16]. In addition, since the dosing schedule of AGs has been shown to be an important determinant in its renal toxicity [1_5,17], the effects of AK on the kidney given as a single injection or as two divided injections were compared. MATERIALS AND METHODS
Animal treatments In a dose-response study, male Hartley guinea pigs (2X&300 g) (Charles River, Montreal, Canada) were injected subcutaneously (s.c.) with amikacin sulfate (Bristol Myers Canada, Belleville, Canada) at doses of 0 (saline-treated), 100, 200 or 400 mg/kg body wt. per day for 10 days. Animals were given the specified dose in either a single injection at 8:00 a.m. or the total daily dose was divided and administered in two injections, one at 8:00 a.m. and one at 4:00 p.m. Each dose group consisted of six
103
animals. Blood samples were collected from the toe of each animal 1 day before AK treatment was started, on the fifth day of AK treatment and at the time of death. The plasma was frozen at -20°C to await analysis of blood urea nitrogen (BUN). BUN was measured using the automated Abbott VP System (Abbott Laboratories Ltd., Mississauga, Canada). Animals were killed by cervical dislocation between 9:00 a.m. and 10:00 a.m. the day following the last dose of AK and the kidneys immediately removed. A section of the left kidney was placed in buffered formalin to await sectioning for light microscopy, while the cortex from the remainder of the left kidney was separated from the medulla and frozen at -80°C for future AK analysis. The right kidney was immediately prepared.for organic ion transport studies. In the time-response study, guinea pigs were injected with AK at a dose of 200 mg/kg body wt. (administered as two equal, S.C.injections per day) for 2, 5,7, 8 or 10 days (six animals/group). The morning following the final injection, the animals were killed and the kidneys were immediately removed. The left kidney cortex was frozen for future AK analysis, while the right kidney was immediately prepared for organic ion transport measurement. Urinalysis Three days prior to any treatment, the guinea pigs were placed in metabolic cages and urine samples were collected daily over sodium azide. At the end of the 24-h collection period, urine volume was recorded and 1 ml sample of each urine was applied onto a Sephadex G-25M PD-10 column (Pharmacia LKB, Uppsala, Sweden). The urine sample was eluted through the column using saline (0.9% NaCl) and the fractions containing creatinine and the enzymes, N-acetyl-/?-D-glucosaminidase (NAG) and alkaline phosphatase (ALP) were collected and stored at 4°C to await analysis. Within 3 days of sample collection (these urinary components were found to be stable over this period of time), creatinine, NAG and ALP were analyzed using the automated Abbott VP system. The concentrations of creatinine, NAG and ALP in the urine collected during the 3 days prior to treatment represented control values for each animal. Creatinine and enzyme excretion during AK treatment were expressed as a percentage of the control values. AK measurement AK was extracted from renal cortical tissue using the technique of Giuliano et al. [ 181. In brief, a weighed sample of kidney cortex was homogenized in a sodium phosphate buffer (10 mM, pH 7.4) and the resultant homogenate was centrifuged at 1200 x g in a Sorval RC-5B refrigerated superspeed centrifuge. The supernatant was collected and the pellet was re-homogenized in another 4 ml of buffer to which 1 ml of trichloroacetic acid (50% w/v) was added. The final homogenate was added to the previous supernatant which was then centrifuged and the pH of the final supernatant was then adjusted to 7.4 using NaOH (1 M). The samples were analyzed in duplicate using an AK radioimmunoassay (RIA) kit (Coat-A-Count Amikacin, Diagnostics
104
Products Co., Los Angeles, CA). There was no interference supematants prepared from control (saline-treated) animals.
with the assay from
Orgaplic ion transport
Transport of the organic ions, p-aminohippuric acid (PAH) and tetraethylammoniurn bromide (TEA) into renal cortical slices was measured according to Smith et al. [ 191. Immediately after a guinea pig was killed, the right kidney was removed and slices of approx. 0.5 mm thickness were cut from the tissue using a Stadie Riggs tissue slicer (Thomas Scientific, Swedesborg, NJ). Slices were incubated for 90 min at 25°C under a gas phase of 100% oxygen in medium containing PAH (7.4 x low5 M) (Sigma Chemical Co., St. Louis, MO) and [14C]TEA (1 x IO-’ M) (New England NuclearDuPont, Mississauga, Canada). Following incubation, slices were removed and concentrations of PAH and TEA in the slices and incubation medium were quantitated. During the standardization of PAH transport in guinea pig slices, it was observed that only about 22% or the PAH added to the renal slice system was recovered following incubation. It was subsequently determined that deacetylation of the slice and medium samples using the method of Carpenter and Mudge (who originally developed this method for mouse renal slice samples) was required to recover all of the PAW in the assay system. The content of [14C]TEA in the renal slices and incubation media was determined by radioactive counting of samples in a Beckman LS3800 (Beckman Instruments Inc., Fullerton, CA) scintillation counter. The transport of PAH and TEA was expressed as a slice/medi~ ratio (S/M) which represents the ratio between the amount of ion taken up into the tissue slice and the amount of ion remaining in the incubation medium. Histopathology
Kidney samples were sectioned, then stained with haematoxylin and eosin (H&E) for light microscopy. Dilation, cell necrosis and regeneration of renal tubules were graded by a pathologist who had no prior knowledge of the animal’s treatment. Morphological changes that were considered attributable to treatment were graded as being absent (-), mild (+) or moderate (++). Statistics
Two sample comparisons were made with Student’s t-test. When multiple comparisons were made, values were analyzed with ANOVA, then subjected to the StudentNeuman-Keuls test as required. A level of P c 0.05 was accepted as being significant. RESULTS
Dose-response study
AK accumulated in the kidney cortex in a dose-related manner (Fig. 1). At the two lowest doses (100 and 200 mg/kg body wt.), significantly more AK accumulated in the
105
‘;i ?? ;i 0
2500 ONE INJECTION 2000
TWO INJECTIONS
E 2
1500
,” g
1000
a: 5:
500
!Ij 0 200
100
DOSE
fmg/kg
400
b.wt.1
Fig. 1. Concentrations of amikacin (AK) in guinea pig kidney cortex following 10 days of S.C.injection of AK at doses of 100, 200 and 400 mg/kg body wt. The total daily dose of AK was given either as a single injection or as two equally divided injections. Values represent the mean f SE, n = 6. * Si~~~antIy different from the one-injection group (P < 0.05).
kidney cortex when the total daily dose was given as two injections versus a single injection. At the dose of 400 mg/kg, there was no significant difference in the renal a~umulation of AK between the single injection and the two injections. BUN levels did not change in animals receiving doses of 100 and 200 mg/kg body wt. even after 10 days of treatment. At the 400 mg dose level, BUN was significantly increased from the pretreatment level of 16.3 mg/dl + 1.9 to 29.7 + 11.5 after 10 days treatment. In the saline-treated guinea pigs, control values for the urinary excretion of creatinine, NAG and, ALP were 5.58 + 0.78 mgldl, 2.14 + 1.24 IU124 h and 0.17 + 0.08 IU/24 h, respectively. The urinary excretion of creatinine and the enzymes did not significantly change during the course of the IO-day saline injection period (data not shown). Urinary creatinine excretion was unaffected by AK treatment. In contrast, at all dose levels, the excretion of both NAG and ALP were significantly increased above control values within I to 3 days of the commencement of AK injections (Figs. 2-4). Although the urinary excretion of NAG was significantly increased during the course of AK injection, due to intragroup variations and large daily fluctuations in the urinary excretion of this enzyme, no trend in NAG excretion could be discerned. Large variations in the urinary excretion of ALP during AK injection also occurred at all doses studied. A distinct peak in ALP excretion was, however, observed with the two-injection dose regimen of AK (Figs. 2B, 3B and 4B). ALP excretion appeared to peak on days 10,9 and 6 at the dose levels of 100, 200 and 400 mg, respectively. Uptake of the cation TEA into renal cortical slices prepared from AK-treated guinea pigs was inhibited in a dose-dependent manner (Fig. 5B). At the doses of 100
106
* -
ONE INJECTION
G---O
TWO INJECTIONS
,
o!
I
0
1
,
,
I
,
I
I
I
,
,
2
3
4
5
6
7
8
9
10
DAYS
1600 -6 1400 -
ONE INJECTION
A---d
TWO
3 2
1200 -
ii
1000 -
INJECTIONS
l
8 600 t; 600 ii! LL
400 -
;i
0, 0
,
,
I
I
,
I
(
I
,
1
1
2
3
4
5
6
7
6
9
10
DAYS
Fig. 2. Urinary excretion profiles of (A) N-acetyl+D-glucosaminidase (NAG) and (B) alkaline phosphatase (ALP) during treatment with AK (100 mg/kg body wt.). The total daily dose of AK was given either as a single injection or as two equally divided injections. Values represent the mean + SE, N = 6. * Significantly different from predose value (100%) (P < 0.05).
and 200 mg, inhibition of TEA was greater when the daily dose of the drug was administered in two injections versus the same dose given in a single injection. As previously mentioned, hydrolysis of the renal slice samples was required prior to measurement of PAH due to the ability of the guinea pig kidney to acetylate PAH. Without prior hydrolysis, only 2 1.8% t 4.9 of the PAH added to the renal slice incubation system was recovered and the S/M ratio was 4.19 + 1.99. Following the deacetylation of the samples, 90-100% of the PAH was recovered and the control S/M ratio was 13.86 + 2.26. Transport of the organic anion, PAH did not appear to be as sensitive as TEA to the effects of AK. Uptake was decreased at doses of 200 mg and greater but there was no difference between the one- or two-injection regimens for any of the doses.
107
1000
z 0
A
-
ONE INJECTION
*---o
TWO INJECTIONS
600
F g
b
600
400
n z
200
0 0
1
2
3
4
5
6
7
6
9
10
DAYS
1600 -6 1600 3 0
1400
-
F z
1200
-
8
1000
-
8
-
ONE INJECTION
&---A
TWO INJECTIONS
l
l
600 -
?J
600
-
fj <
400
-
200 0 0
I 1
I 2
I 3
I1 4
5
I 6
I,
7
8
I, 9
10
DAYS Fig. 3. Urinary exctetion profiles of (A) NAG and (B) ALP during treatment with AK (200 mg/kg body wt.). The total daily dose of AK was given either as a single injection or as two equally divided injections. Values represent the mean f SE, n = 6. * Significant different from predose value (100%) (P < 0.05).
Time-response study Because the dose of 200 mg/kg body wt. was high enough to cause proximal tubular damage but did not alter BUN, this dose was chosen for the remainder of the study to determine the time course of AK-induced effects on organic ion transport. The longer the dosing period, the greater the renal concentration of AK (Table I). PAH uptake was found to be reduced only on the tenth day of treatment (Table I). The uptake of TEA was significantly reduced by the eighth day of treatment and remained depressed through to day 10 (Table I). Histopathology Mild interstitial nephritis and glomerulonephritis were present in many animals regardless of treatment. Tubular dilation, necrosis and regeneration were absent in
108 1000
‘3 0
+
-A
800 -
-
ONE INJECTION
O---O
TWO INJECTIONS
F $
600
0, 0
(
/
(
(
,
,
,
,
/
,
1
2
3
4
5
6
7
6
9
10
6
7
6
9
10
DAYS
-
0
ONE INJECTION
1
2
3
4
5
DAYS
Fig. 4. Urinary excretion profiles of (A) NAG and (B) ALP during treatment with AK (400 mgkg body wt.). The total daily dose of AK was given either as a single injection or as two equally divided injections. Values represent the mean a SE, n = 6. * Significantly different from predose value (100%) (P < 0.05).
the control animals and were more prominent depending on treatment and dose regimen (Table II). At the dose of 100 mg/kg body wt., some tubular dilation was observed and there was no difference between the two dosing regimens. At 200 mglkg body wt., tubular dilation was present in both dosing regimen groups. However, tubular necrosis and regeneration were present only in the two-injection group. At 400 mgikg body wt., tubular dilation was most prominent, while regeneration and necrosis were comprable to other dosages. Both dosing regimens appeared to be equally affected at this dose. DISCUSSION
In response to concerns about the absence of reliable tests to detect AG-induced
109
CONTROL
a
ONE INJECTION
a
TWO INJECTIONS
100
200
400
(
ONE INJECTION
m
TWO INJECTIONS
Fig. 5. Uptake of (A) p-aminohippuric acid (PAH) and (B) tetraethylammonium (TEA) in renal cortical slices prepared from kidneys of guinea pigs injected for 10 days with AK at doses of 100,200 and 400 mg/kg body wt. The total daily dose of AK was given either as a single injection or as two equally divided injections. Values represent the mean f SE, n = 6. * Significantly different from control (P c 0.05). ‘+‘Significantly different from the one-injection group (P < 0.05).
renal tubular damage, a number of investigators have searched for more sensitive markers of nephrotoxicity. Although elevations in the urinary excretion of marker enzymes following AG treatment have been demonstrated [7-141, the usefulness of enzymuria as a prognostic device to predict the degree of renal tubular damage during the course of AG therapy has not yet been established. Examination of AG-induced enzymuria in earlier animal studies have generally been limited to a small number of doses which were often quite high and resulted in extensive kidney damage. In addition, usually only a single dosing regimen was used and urinary enzyme values were not measured on a daily basis. Finally, although several studies have examined gentamicin (GM)-induced enzymuria, fewer studies have looked at enzymuria following
110 TABLE I RENAL CORTICAL CONCENTRATIONS OF AMIKACIN (AK) AND THE TRANSPORT OF TETRAETHYLAMMONIUM (TEA) ANDp-AMINOHIPPURIC ACID (PAH) INTO RENAL CORTICAL SLICES FOLLOWING IN VIVO TREATMENT WITH AK” Day
AK &gig)
0
2 5 I 8 10
NW
564.75 i 52.30 804.24 ? 35.90 913.30 + 46.46 790.20 t 110.06 1155.0 t 154.46
TEA (S/M)b
PAH (S/M)h
9.25 8.70 8.99 8.48 6.25 5.74
13.86 rf:0.92 14.68 t 0.76 15.49 t 0.81 13.65 It 0.57 12.90 + 1.22 9.74 t 1.03*
* 0.44 t 0.36 f 0.10 & 0.52 f 0.81* f 0.37”
a All animals except those in the day 0 group were treated with AK at a dose of 200 mg/kg body wt. given in two injections per day (values represent the mean t SE, n = 6). ‘Represents the ratio between the amount of ion taken up per gram of tissue slice and the amount of ion remaining in the incubation media. ‘Not determined. *Significantly different from day 0 (P <: 0.05).
exposure to AK. Therefore a study was undertaken which assessed the nephrotoxicity of AK in guinea pigs utilizing several doses and different dosage regimens. As previously mentioned, organic ion transport has been demonstrated to be sensitive to the nephrotoxic effects of the AGs [15,16] and were, therefore, monitored in the present study. Since studies where organic ion transport is measured in guinea pigs are rare, there is a paucity of information regarding uptake measurement in this species. In the present investigation it was observed that, similar to the mouse [20], the guinea pig kidney rapidly acetylates PAH thereby requiring deacetylation of samples prior to analysis. The uptake of PAH was found not to be particularly sensitive to the effects of AK which confirms findings of earlier studies with other AGs in other species [l&21]. In contrast to PAH, TEA transport was found to be more sensitive to AK-induced nephrotoxicity and its uptake was reduced in a dose-dependent manner. This inhibition occurred at doses which did not alter BUN concentrations re-emphasizing the relative insensitivity of BUN as an indicator of AK-induced renal damage. One of the earliest events in the pathogenesis of AG-induced nephrotoxicity is the disruption of the lysosomal system with changes in lysosomal size, number and appearance 1221preceding other overt signs of tubular cell damage [23]. In both the present study and earlier reports by others [7,12,14], urinary NAG concentrations were elevated within l-3 days of the commencement of AG treatment. The early appearance of this lysosomal enzyme in the absence of brush border membrane damage is not surprising since, due to its solubility, NAG can be released from disrupted lysosomes across the brush border membrane into the tubular lumen [5]. In a study by Gibey et al. [12], it was proposed that a relationship between the
111
TABLE II CHANGES IN KIDNEY TUBULAR MORPHOLOGY TREATMENT WITH AMIKACIN (AK) Dose of AK
Control 100 (1) log (2) 200 (1) 200 (2) 400 (1) 400 (2) a The total daily bMorphological ’ Morphological d Mo~hological
Tubular dilation _b +c + + + ++d ++
IN GUINEA PIGS AFTER 10 DAYS OF
Tubular cell necrosis
Tubular regeneration
_
-
+ + -
+ + +
+ -
dose of AK was administered either in one injection (1) or two injections (2). changes absent. changes mild. changes moderate.
nephrotoxicity of GM and the initial urinary NAG activity may exist and that the degree of NAG response during the first 10 days of treatment may serve as a prognostic factor. It was further suggested by Donta and Lembke [I l] that the rate of increase to peak NAG excretion and the degree of toxicity could be correlated in patients receiving GM. In contrast to these studies, the results of the present study fail to show such relationships. Due to large intragroup variations, daily fluctuations and the absence of any distinct urinary excretion pattern for NAG, there does not appear to be any quantitative relationship between AK-induced renal damage and enzymuria. Large day-to-day variations in the urinary excretion of NAG in newborns were also reported by Adelman et al. [24] who concluded that because of these variations, NAG would not be particularly useful in predicting subsequent changes in renal function caused by AGs. In another clinical study by Fujita et al. 1251,it was also concluded that due to the large variation in the urinary NAG response in patients during AG therapy, the degree of nephrotoxicity of AG as a function of urinary NAG levels could not be shown. The sensitivity of the brush border membrane enzyme, ALP to AK-induced renal damage was also examined, however, similar to NAG, large daily variations in the urinary excretion of ALP were observed during treatment with AK in the one-injection groups. In contrast, with the two-injection regimen, a distinct peak in ALP excretion was observed. It was also observed that, for this dosing regimen, the higher the drug dose, the earlier the appearance of the peak in ALP excretion. In the 200 mg dose group (two injections/day regimen), the appearance of this peak (on day 9) approximately coincided with the decrease in TEA uptake which occurred by the eighth day of treatment. Kaloyanides 1261also reported sharp increases in enzymuria during the course of GM injection to rats which coincided with the onset of proximal
112
tubular cell necrosis. Thus, the peak urinary excretion of the brush border membrane enzyme, ALP in the two-injection group may reflect the onset of tubular cell necrosis. Why similar peaks in the excretions of NAG and ALP (in the one-injection regimen) with different doses of AK treatment did not occur is unknown but emphasizes the difficulty in utilizing enzymuria as a quantitative marker of AK-induced renal damage. The dosing schedule is known to be an important factor in the nephrotoxicity of AGs. Administration of GM in divided doses has been shown to result in greater damage to the kidney than the administration of the same dose of the drug in a single injection [15,27,28] presumably due to a greater renal accumulation of the drug following the multiple dosing [17,29-311. Because the renal uptake mechanism for AGs is a saturable process, the efficiency of uptake decreases as plasma concentrations increase [32]. Therefore, following multiple dosing, a larger proportion of the drug can be taken up into the kidney since plasma levels will be lower after each divided dose than those attained following administration of the drug in a single injection. In the present study, at the two lower doses, administration of AK in two injections per day resulted in higher drug concentrations in the kidney. There also appeared to be greater renal damage with this regimen since there was a greater reduction in TEA transport as compared to the one-injection regimen. In addition, at the dose of 200 mgikg body wt., the two-injection regimen also resulted in greater changes in tubular morphology than the one-injection regimen. The difference in toxicity between the two different dosing regimens could not be discerned from the enzymuria data again due to large intragroup variations, daity ~uctuations and an absence of discernable trends in the urinary excretion of NAG and ALP. The difference in toxicity between the two dosing regimens was not observed at the 100 mg dose levels. However, at this lower dose, morphological changes observed in both dosing regimen groups were mild and limited to tubular dilation. At the dose of 400 mgikg body wt., division of the dose did not result in either a greater accumulation of the drug in the kidney or in greater nephrotoxicity. It is likely that at this dose, even with the divided dose, saturation of the uptake mechanism had occurred and therefore renal accumulation of the drug was limited. In conclusion, results from this investigation into the nephrotoxicity of AK suggest that, despite significant increases in the urinary excretions of NAG and ALP following AK administration, these enzymes do not appear to be useful quantitative or prognostic markers of AG-induced renal damage due to the difficulty in establishing a relationship between enzymuria and kidney damage.
The authors wish to thank Messrs. C. Paul, Solomonraj, K. Kittle and L. Samure for their excellent technical assistance.
113 REFERENCES 1 Meyer, R.D. (1981) Amikacin. Ann. Intern. Med. 95,32&332. 2 Gingell, J.C. and Waterworth, P.M. (1968) Dose of gentamicin in patients with normal renal function and renal impairment. Br. Med. J. 2, 19-22. 3 Price, R.G. (1982) Urinary enzymes, nephrotoxicity and renal disease. Toxicology 23,99-134. 4 Palla, R., Marchitiello, M., Tuoni, M., Cirami, C., Giovannini, L., Bertelli, A.A.E. and Bertelli, A. (1985) Enzymuria in aminoglycoside-induced kidney damage. Comparative study of gentamicin, amikacin, sisomicin and netilmicin. Int. J. Clin. Pharmacol. Res. 5, 351-355. 5 Pipemo, E. (1981) Detection of drug induced nephrotoxicity with urinanalysis and enzymuria assessment. In: J.B. Hook (Ed.), Toxicology of the Kidney, Raven Press, New York, pp. 31-55. 6 Smith, J.H. and Hook, J.B. (1982) Experimental nephrotoxicity in vivo. In: P.H. Bach, R.W. Bonner, J.W. Bridges and E.A. Lock (Eds.), Nephrotoxicity Assessment and Pathogenesis, John Wiley and Sons, Chichester, pp. 117-127. 7 Garry, F., Chew, D.J. and Hoffsis, G.F. (1990) Enzymuria as an index of renal damage in sheep with induced aminoglycoside nephrotoxicosis. Am. J. Vet. Res. 51, 428432. 8 Luft, F.C., Pate], V., Yum, M.N., Patel, B. and Kleit, S.A. (1975) Experimental aminoglycoside nephrotoxicity. J. Lab. Clin. Med. 86, 213-220. 9 Patel, V., Luft, F.C., Yum, M.N., Patel, B., Zeeman, W. and Kleit, S.A. (1975) Enzymuria in gentamitin-induced kidney damage. Antimicrob. Agents Chemother. 8, 364369. 10 Wellwood, J.M., Lovell, D., Thompson, A& and Tighe, J.R. (1976) Renal damage caused by gentamicin: A study of the effects on renal morphology and urinary enzyme excretion. J. Pathol. 118, 171-182. 11 Donta, S.T. and Lembke, L.A. (1985) Com~rative effects of gentamicin and tobr~ycin on excretion of ~-acetyl-~-D-glucos~inida~. Antimicrob. Agents Chemother. 28, 500-503. 12 Gibey, R., Dupond, J.L., Alber, D., Lecorte des Floris, R. and Henry, J.C. (1981) predictive value of urinary N-acetyl-/I-n-glucosaminidase and p-2-microglobulin (82 M) in evaluating nephrotoxicity of gentamicin. Clin. Chim. Acta. 116,25-34. 13 Mondorf, A.W., Brier, J., Hindus, J., Scherberich, J.E., Mackenrodt, G., Shah, P.M., Stille, W. and Schoeppe, W. (1978) Effect of aminoglycosides on proximal tubular membranes of the human kidney. Eur. J. Clin. Pharmacol. 13, 133-142. 14 Palla, R., Paternoster, G., Galigani, P., Giovanni, L., Bertelli, A.A.E., Romano, M.R., Alessandri, M.G. and Bertelli, A. (1987) Comparative effects of gentamicin, amikacin and dactimicin on excretion of N-acetyl-/+glucosaminidase (NAG) and kidney histological pattern in rats. Drugs Exp. Clin. Res. 13,751-756. 15 Bennett, W.M., Plamp, C.E., Gilbert, D.N., Parker, R.A. and Porter, G.A. (1979) The influence of dosage regimen on ex~rimental gentamicin nephrotoxicity: Dissociation of peak serum levels from renal failure. J. Infect. Dis. 140, 576580. 16 Kluwe, W.M. and Hook, J.B. (1978) Functional nephrotoxicity of gentamicin in the rat. Toxicol. Appl. Pharmacol. 45,163-l 75. 17 Reiner, N.E., Bloxham, D.D. and Thompson, W.L. (1978) Nephrotoxicity of gentamicin and tobramytin given once daily or continuously in dogs. J. Antimicrob. Chemother. 4, 855101. 18 Giuliano, R.A., Verpooten, G.A., Pollet, DE., Verbist, L., Scharpe, S.L. and DeBroe, M.E. (1984) Improved procedure for extracting aminoglycosides from renal cortical tissue. Antimicrob. Agents Chemother. 25, 783-784. 19 Smith, J.H., Braselton, W.E., Tonsager, S.R., Mayor, G.H. and Hook, J.B. (1982) Effects of vanadate on organic ion accumulation in rat renal cortical slices. J. Pharmacol. Exp. Ther. 220, 540-546. 20 Carpenter, H.M. and Mudge, G.H. (1980) Uptake and acetylation of p-aminohippurate by slices of mouse kidney cortex. J. Pharmacol. Exp. Ther. 213,3X)-354. 21 Bennett, W-M., Plamp, C.E., Parker, R.A., Gilbert, D.N.. Houghton, D.C. and Porter, G.A. (1980)
114
Alterations 95, 32239.
in organic
22 Houghton,
ion transport
D.C., Hartnett,
electron
microscopic
23 Tulkens,
nisms and perspectives.
of gentamicin
K., Murayama,
26 Kaloyanides, tex. Fund. 27 Aronoff,
M., Porter,
nephrotoxicity
in the rat. J. Lab. Clin. Med.
G. and Bennett,
W. (1976) A light and
in rats. Am. J. Physiol.
studies on nephrotoxicity
of aminoglycosides
F. and Rubis, T. (1987) A controlled
in the neonate.
cosides on excretion
nephrotoxicity
82, 589-612. at low doses: Mecha-
Am. J. Med. 80, 5514.
R.D., Wirth,
and amikacin
by gentamicin
M., Campbell-Boswell,
analysis
P.M. (1986) Experimental
24 Adelman, 25 Fujita,
induced
T., Kawamura,
M. and Narita,
of N-acetyl-B-o-glucosaminidase.
G.R., Pottratz,
K. (1987) Comparison Int. J. Clin. Pharm.
G.J. (1984) Aminoglycoside-induced Appl. Toxicol.
study of the nephrotoxicity
of mezlocillin
Am. J. Dis. Child 141, 1175-l 178.
functional
of the effect of aminoglyRes. 7, 4777480.
and biochemical
defects in the renal cor-
4, 930-943. S.T., Brier, M.E., Walker,
(1983) Aminoglycosideaccumulation
N.E., Fineberg,
N.S., Glant,
kinetics in rat renal parenchyma.
M.D. and Luft, F.C.
Antimicrob.
Agents Chemother.
23, 74478. 28 Powell,
S.H.,
Groden,
D.L.,
Thompson, Jacobs,
W.L., M.R.,
continuous
aminoglycoside
netilmicin
and tobramycin.
Luthe,
Discenna,
dosing:
M.A., A.O.,
Efficacy
Stern, Cash,
and toxicity
R.C., H.A.
Grossniklaus,
and Klinger,
in animals
D.A.,
R.A., Verpooten,
kinetics of aminoglycosides 31 Verpooten, dosing 32 DeBroe,
G.A.,
decreases
Giuliano,
vs.
J. Infect. Dis. 147, 918-932.
G.A., Verbist,
L., Wedeen,
R.P. and DeBroe,
in the kidney cortex of rats. J. Pharmacol. R.A.,
renal accumulation
M.E., Giuliano,
D.D.,
and clinical studies of gentamicin,
29 Giulano, R.A., Verpooten, G.A. and DeBroe, M.E. (1986) The effect of dosing cortical accumulation of aminoglycosides in rats. Am. J. Kidney Dis. 8, 297-303. 30 Giuliano,
Bloxham,
J.D. (1983) Once daily
Verbist,
L., Eestermans,
of gentamicin
R.A. and Verpooten,
on kidney
M.E. (1986) In vivo uptake
Exp. Ther. 236, 470-475.
G. and DeBroe,
and netilmicin.
strategy
M.E. (1989) Once-daily
Clin. Pharmacol.
G.A. (1989) Aminoglycoside
Ther. 45, 22-27.
nephrotoxicity:
nism and prevention. In: A. Ameriao, P. Coratelli, V.M. Cambese and S.G. Massry Systemic Diseases and the Kidney, Plenum Publishing, New York, pp. 2333245.
Mecha-
(Eds), Drugs,
101
TOXLET 02761
Evaluation of enzymuria as an indicator of amikacininduced renal damage in guinea pigs
Catherine A.M. Suzuki’, Barry H. Thomas and Ruedi Mueller’ Drug Toxicology Division, Bureau of Drug Research, Health Protection Branch, Sir F.G. Banting Research Centre, Tunney’s Pasture, Ottawa, Ontario (Canada)
(Received 23 January 1992) (Accepted 3 April 1992) Key words: Amikacin; Enzymuria; Nephrotoxicity
SUMMARY Guinea pigs were injected subcutaneously for 10 days with amikacin (AK) at a dose of 0, 100,200 or 400 mg/kg body wt. per day. The total daily dose was administered in either a single injection or divided equally and given as two daily injections. After the 10 days of AK treatment, uptake of the organic cation, tetraethylammonium (TEA) into renal cortical slices was inhibited in a dose-related manner. Changes in renal tubular morphology also increased with higher doses. The urinary excretion of the enzymes, N-acetyl/?-o-glucosaminidase (NAG), and alkaline phosphatase (ALP) significantly increased during the course of AK treatment, however, due to the large intragroup variability, the daily fluctuations and the absence of any distinct trends in urinary enzyme excretion it was difficult to establish a dose-relationship between AK-induced renal damage and the resultant enzymuria. At doses of 100 and 200 mg/kg body wt., the two-injection regimen resulted in the greater renal accumulation of AK and damage as reflected by a greater inhibition of TEA uptake and greater changes in renal tubular morphology. In contrast, this difference in toxicity could not be detected with enzymuria again due to the large intragroup variability and the absence of discernable excretion patterns of NAG and ALP. Thus, neither NAG nor ALP appear to be suitable quantitative markers of AK-induced nephrotoxicity.
‘Present address: Toxicology Research Division, Bureau of Chemical Safety, Health Protection Branch, Sir F.G. Banting Research Centre, Tunney’s Pasture, Ottawa, Ontario, Canada KlA 0L2. Correspondence to: C.A.M. Suzuki, Toxicology Research Division, Bureau of Chemical Safety, Health Protection Branch, Sir F.G. Banting Research Centre, Tunney’s Pasture, Ottawa, Ontario, Canada KlA OL2.
102
INTRODUCTION
The aminoglycoside (AG) antibiotics represent the first line of defence against a number of Gram-negative bacterial infections. Unfortunately, use of AGs may be limited due to a number of potential side effects including ototoxicity and nephrotoxicity [l]. Although the renal damage caused by these drugs is generally reversible, impairment of renal function during the course of therapy may, however, result in the reduction of AG clearance since 80-90% of these drugs are eliminated from the body via the kidney [l]. Reduced clearance may, in turn, result in sustained, elevated plasma drug concentrations [2], resulting in the accumulation of the drug to toxic levels in other sensitive target organs such as the ear. Consequently, monitoring the renal status of a patient during the course of AG treatment is of great importance. Currently, serum creatinine and/or blood urea nitrogen levels are used to monitor the renal function of patients undergoing AG therapy. However, these parameters are relatively insensitive markers of renal damage and become significantly altered only after considerable damage to the kidney has occurred [3]. Recently the measurement of enzymuria (renal-derived enzymes in the urine) has received some attention as a potential indicator of AG-induced nephrotoxicity [3-61. Although normally present only in low concentrations in the urine, following damage to the kidney, concentrations of certain enzymes in the urine can significantly increase. In some cases, their appearance in the urine may precede changes in traditional renal function tests [6]. Although the occurrence of enzymuria during gentamicin treatment has been reported with both experimental animals [7-lo] and patients receiving AGs [l l-141, the quantitative relationship between enzymuria and renal damage has not yet been well defined. Thus, the present study was undertaken to further examine this relationship using the AG, amikacin (AK). Amikacin-induced renal damage was assessed by measuring the urinary excretions of N-acetyl-/?-o-glucosaminidase (NAG) and alkaline phosphatase (ALP). To evaluate the usefulness of enzymuria as a quantitative marker of renal damage, organic ion transport in renal cortical slices was also measured due to the sensitivity of ion transport to AG-induced nephrotoxicity [15,16]. In addition, since the dosing schedule of AGs has been shown to be an important determinant in its renal toxicity [1_5,17], the effects of AK on the kidney given as a single injection or as two divided injections were compared. MATERIALS AND METHODS
Animal treatments In a dose-response study, male Hartley guinea pigs (2X&300 g) (Charles River, Montreal, Canada) were injected subcutaneously (s.c.) with amikacin sulfate (Bristol Myers Canada, Belleville, Canada) at doses of 0 (saline-treated), 100, 200 or 400 mg/kg body wt. per day for 10 days. Animals were given the specified dose in either a single injection at 8:00 a.m. or the total daily dose was divided and administered in two injections, one at 8:00 a.m. and one at 4:00 p.m. Each dose group consisted of six
103
animals. Blood samples were collected from the toe of each animal 1 day before AK treatment was started, on the fifth day of AK treatment and at the time of death. The plasma was frozen at -20°C to await analysis of blood urea nitrogen (BUN). BUN was measured using the automated Abbott VP System (Abbott Laboratories Ltd., Mississauga, Canada). Animals were killed by cervical dislocation between 9:00 a.m. and 10:00 a.m. the day following the last dose of AK and the kidneys immediately removed. A section of the left kidney was placed in buffered formalin to await sectioning for light microscopy, while the cortex from the remainder of the left kidney was separated from the medulla and frozen at -80°C for future AK analysis. The right kidney was immediately prepared.for organic ion transport studies. In the time-response study, guinea pigs were injected with AK at a dose of 200 mg/kg body wt. (administered as two equal, S.C.injections per day) for 2, 5,7, 8 or 10 days (six animals/group). The morning following the final injection, the animals were killed and the kidneys were immediately removed. The left kidney cortex was frozen for future AK analysis, while the right kidney was immediately prepared for organic ion transport measurement. Urinalysis Three days prior to any treatment, the guinea pigs were placed in metabolic cages and urine samples were collected daily over sodium azide. At the end of the 24-h collection period, urine volume was recorded and 1 ml sample of each urine was applied onto a Sephadex G-25M PD-10 column (Pharmacia LKB, Uppsala, Sweden). The urine sample was eluted through the column using saline (0.9% NaCl) and the fractions containing creatinine and the enzymes, N-acetyl-/?-D-glucosaminidase (NAG) and alkaline phosphatase (ALP) were collected and stored at 4°C to await analysis. Within 3 days of sample collection (these urinary components were found to be stable over this period of time), creatinine, NAG and ALP were analyzed using the automated Abbott VP system. The concentrations of creatinine, NAG and ALP in the urine collected during the 3 days prior to treatment represented control values for each animal. Creatinine and enzyme excretion during AK treatment were expressed as a percentage of the control values. AK measurement AK was extracted from renal cortical tissue using the technique of Giuliano et al. [ 181. In brief, a weighed sample of kidney cortex was homogenized in a sodium phosphate buffer (10 mM, pH 7.4) and the resultant homogenate was centrifuged at 1200 x g in a Sorval RC-5B refrigerated superspeed centrifuge. The supernatant was collected and the pellet was re-homogenized in another 4 ml of buffer to which 1 ml of trichloroacetic acid (50% w/v) was added. The final homogenate was added to the previous supernatant which was then centrifuged and the pH of the final supernatant was then adjusted to 7.4 using NaOH (1 M). The samples were analyzed in duplicate using an AK radioimmunoassay (RIA) kit (Coat-A-Count Amikacin, Diagnostics
104
Products Co., Los Angeles, CA). There was no interference supematants prepared from control (saline-treated) animals.
with the assay from
Orgaplic ion transport
Transport of the organic ions, p-aminohippuric acid (PAH) and tetraethylammoniurn bromide (TEA) into renal cortical slices was measured according to Smith et al. [ 191. Immediately after a guinea pig was killed, the right kidney was removed and slices of approx. 0.5 mm thickness were cut from the tissue using a Stadie Riggs tissue slicer (Thomas Scientific, Swedesborg, NJ). Slices were incubated for 90 min at 25°C under a gas phase of 100% oxygen in medium containing PAH (7.4 x low5 M) (Sigma Chemical Co., St. Louis, MO) and [14C]TEA (1 x IO-’ M) (New England NuclearDuPont, Mississauga, Canada). Following incubation, slices were removed and concentrations of PAH and TEA in the slices and incubation medium were quantitated. During the standardization of PAH transport in guinea pig slices, it was observed that only about 22% or the PAH added to the renal slice system was recovered following incubation. It was subsequently determined that deacetylation of the slice and medium samples using the method of Carpenter and Mudge (who originally developed this method for mouse renal slice samples) was required to recover all of the PAW in the assay system. The content of [14C]TEA in the renal slices and incubation media was determined by radioactive counting of samples in a Beckman LS3800 (Beckman Instruments Inc., Fullerton, CA) scintillation counter. The transport of PAH and TEA was expressed as a slice/medi~ ratio (S/M) which represents the ratio between the amount of ion taken up into the tissue slice and the amount of ion remaining in the incubation medium. Histopathology
Kidney samples were sectioned, then stained with haematoxylin and eosin (H&E) for light microscopy. Dilation, cell necrosis and regeneration of renal tubules were graded by a pathologist who had no prior knowledge of the animal’s treatment. Morphological changes that were considered attributable to treatment were graded as being absent (-), mild (+) or moderate (++). Statistics
Two sample comparisons were made with Student’s t-test. When multiple comparisons were made, values were analyzed with ANOVA, then subjected to the StudentNeuman-Keuls test as required. A level of P c 0.05 was accepted as being significant. RESULTS
Dose-response study
AK accumulated in the kidney cortex in a dose-related manner (Fig. 1). At the two lowest doses (100 and 200 mg/kg body wt.), significantly more AK accumulated in the
105
‘;i ?? ;i 0
2500 ONE INJECTION 2000
TWO INJECTIONS
E 2
1500
,” g
1000
a: 5:
500
!Ij 0 200
100
DOSE
fmg/kg
400
b.wt.1
Fig. 1. Concentrations of amikacin (AK) in guinea pig kidney cortex following 10 days of S.C.injection of AK at doses of 100, 200 and 400 mg/kg body wt. The total daily dose of AK was given either as a single injection or as two equally divided injections. Values represent the mean f SE, n = 6. * Si~~~antIy different from the one-injection group (P < 0.05).
kidney cortex when the total daily dose was given as two injections versus a single injection. At the dose of 400 mg/kg, there was no significant difference in the renal a~umulation of AK between the single injection and the two injections. BUN levels did not change in animals receiving doses of 100 and 200 mg/kg body wt. even after 10 days of treatment. At the 400 mg dose level, BUN was significantly increased from the pretreatment level of 16.3 mg/dl + 1.9 to 29.7 + 11.5 after 10 days treatment. In the saline-treated guinea pigs, control values for the urinary excretion of creatinine, NAG and, ALP were 5.58 + 0.78 mgldl, 2.14 + 1.24 IU124 h and 0.17 + 0.08 IU/24 h, respectively. The urinary excretion of creatinine and the enzymes did not significantly change during the course of the IO-day saline injection period (data not shown). Urinary creatinine excretion was unaffected by AK treatment. In contrast, at all dose levels, the excretion of both NAG and ALP were significantly increased above control values within I to 3 days of the commencement of AK injections (Figs. 2-4). Although the urinary excretion of NAG was significantly increased during the course of AK injection, due to intragroup variations and large daily fluctuations in the urinary excretion of this enzyme, no trend in NAG excretion could be discerned. Large variations in the urinary excretion of ALP during AK injection also occurred at all doses studied. A distinct peak in ALP excretion was, however, observed with the two-injection dose regimen of AK (Figs. 2B, 3B and 4B). ALP excretion appeared to peak on days 10,9 and 6 at the dose levels of 100, 200 and 400 mg, respectively. Uptake of the cation TEA into renal cortical slices prepared from AK-treated guinea pigs was inhibited in a dose-dependent manner (Fig. 5B). At the doses of 100
106
* -
ONE INJECTION
G---O
TWO INJECTIONS
,
o!
I
0
1
,
,
I
,
I
I
I
,
,
2
3
4
5
6
7
8
9
10
DAYS
1600 -6 1400 -
ONE INJECTION
A---d
TWO
3 2
1200 -
ii
1000 -
INJECTIONS
l
8 600 t; 600 ii! LL
400 -
;i
0, 0
,
,
I
I
,
I
(
I
,
1
1
2
3
4
5
6
7
6
9
10
DAYS
Fig. 2. Urinary excretion profiles of (A) N-acetyl+D-glucosaminidase (NAG) and (B) alkaline phosphatase (ALP) during treatment with AK (100 mg/kg body wt.). The total daily dose of AK was given either as a single injection or as two equally divided injections. Values represent the mean + SE, N = 6. * Significantly different from predose value (100%) (P < 0.05).
and 200 mg, inhibition of TEA was greater when the daily dose of the drug was administered in two injections versus the same dose given in a single injection. As previously mentioned, hydrolysis of the renal slice samples was required prior to measurement of PAH due to the ability of the guinea pig kidney to acetylate PAH. Without prior hydrolysis, only 2 1.8% t 4.9 of the PAH added to the renal slice incubation system was recovered and the S/M ratio was 4.19 + 1.99. Following the deacetylation of the samples, 90-100% of the PAH was recovered and the control S/M ratio was 13.86 + 2.26. Transport of the organic anion, PAH did not appear to be as sensitive as TEA to the effects of AK. Uptake was decreased at doses of 200 mg and greater but there was no difference between the one- or two-injection regimens for any of the doses.
107
1000
z 0
A
-
ONE INJECTION
*---o
TWO INJECTIONS
600
F g
b
600
400
n z
200
0 0
1
2
3
4
5
6
7
6
9
10
DAYS
1600 -6 1600 3 0
1400
-
F z
1200
-
8
1000
-
8
-
ONE INJECTION
&---A
TWO INJECTIONS
l
l
600 -
?J
600
-
fj <
400
-
200 0 0
I 1
I 2
I 3
I1 4
5
I 6
I,
7
8
I, 9
10
DAYS Fig. 3. Urinary exctetion profiles of (A) NAG and (B) ALP during treatment with AK (200 mg/kg body wt.). The total daily dose of AK was given either as a single injection or as two equally divided injections. Values represent the mean f SE, n = 6. * Significant different from predose value (100%) (P < 0.05).
Time-response study Because the dose of 200 mg/kg body wt. was high enough to cause proximal tubular damage but did not alter BUN, this dose was chosen for the remainder of the study to determine the time course of AK-induced effects on organic ion transport. The longer the dosing period, the greater the renal concentration of AK (Table I). PAH uptake was found to be reduced only on the tenth day of treatment (Table I). The uptake of TEA was significantly reduced by the eighth day of treatment and remained depressed through to day 10 (Table I). Histopathology Mild interstitial nephritis and glomerulonephritis were present in many animals regardless of treatment. Tubular dilation, necrosis and regeneration were absent in
108 1000
‘3 0
+
-A
800 -
-
ONE INJECTION
O---O
TWO INJECTIONS
F $
600
0, 0
(
/
(
(
,
,
,
,
/
,
1
2
3
4
5
6
7
6
9
10
6
7
6
9
10
DAYS
-
0
ONE INJECTION
1
2
3
4
5
DAYS
Fig. 4. Urinary excretion profiles of (A) NAG and (B) ALP during treatment with AK (400 mgkg body wt.). The total daily dose of AK was given either as a single injection or as two equally divided injections. Values represent the mean a SE, n = 6. * Significantly different from predose value (100%) (P < 0.05).
the control animals and were more prominent depending on treatment and dose regimen (Table II). At the dose of 100 mg/kg body wt., some tubular dilation was observed and there was no difference between the two dosing regimens. At 200 mglkg body wt., tubular dilation was present in both dosing regimen groups. However, tubular necrosis and regeneration were present only in the two-injection group. At 400 mgikg body wt., tubular dilation was most prominent, while regeneration and necrosis were comprable to other dosages. Both dosing regimens appeared to be equally affected at this dose. DISCUSSION
In response to concerns about the absence of reliable tests to detect AG-induced
109
CONTROL
a
ONE INJECTION
a
TWO INJECTIONS
100
200
400
(
ONE INJECTION
m
TWO INJECTIONS
Fig. 5. Uptake of (A) p-aminohippuric acid (PAH) and (B) tetraethylammonium (TEA) in renal cortical slices prepared from kidneys of guinea pigs injected for 10 days with AK at doses of 100,200 and 400 mg/kg body wt. The total daily dose of AK was given either as a single injection or as two equally divided injections. Values represent the mean f SE, n = 6. * Significantly different from control (P c 0.05). ‘+‘Significantly different from the one-injection group (P < 0.05).
renal tubular damage, a number of investigators have searched for more sensitive markers of nephrotoxicity. Although elevations in the urinary excretion of marker enzymes following AG treatment have been demonstrated [7-141, the usefulness of enzymuria as a prognostic device to predict the degree of renal tubular damage during the course of AG therapy has not yet been established. Examination of AG-induced enzymuria in earlier animal studies have generally been limited to a small number of doses which were often quite high and resulted in extensive kidney damage. In addition, usually only a single dosing regimen was used and urinary enzyme values were not measured on a daily basis. Finally, although several studies have examined gentamicin (GM)-induced enzymuria, fewer studies have looked at enzymuria following
110 TABLE I RENAL CORTICAL CONCENTRATIONS OF AMIKACIN (AK) AND THE TRANSPORT OF TETRAETHYLAMMONIUM (TEA) ANDp-AMINOHIPPURIC ACID (PAH) INTO RENAL CORTICAL SLICES FOLLOWING IN VIVO TREATMENT WITH AK” Day
AK &gig)
0
2 5 I 8 10
NW
564.75 i 52.30 804.24 ? 35.90 913.30 + 46.46 790.20 t 110.06 1155.0 t 154.46
TEA (S/M)b
PAH (S/M)h
9.25 8.70 8.99 8.48 6.25 5.74
13.86 rf:0.92 14.68 t 0.76 15.49 t 0.81 13.65 It 0.57 12.90 + 1.22 9.74 t 1.03*
* 0.44 t 0.36 f 0.10 & 0.52 f 0.81* f 0.37”
a All animals except those in the day 0 group were treated with AK at a dose of 200 mg/kg body wt. given in two injections per day (values represent the mean t SE, n = 6). ‘Represents the ratio between the amount of ion taken up per gram of tissue slice and the amount of ion remaining in the incubation media. ‘Not determined. *Significantly different from day 0 (P <: 0.05).
exposure to AK. Therefore a study was undertaken which assessed the nephrotoxicity of AK in guinea pigs utilizing several doses and different dosage regimens. As previously mentioned, organic ion transport has been demonstrated to be sensitive to the nephrotoxic effects of the AGs [15,16] and were, therefore, monitored in the present study. Since studies where organic ion transport is measured in guinea pigs are rare, there is a paucity of information regarding uptake measurement in this species. In the present investigation it was observed that, similar to the mouse [20], the guinea pig kidney rapidly acetylates PAH thereby requiring deacetylation of samples prior to analysis. The uptake of PAH was found not to be particularly sensitive to the effects of AK which confirms findings of earlier studies with other AGs in other species [l&21]. In contrast to PAH, TEA transport was found to be more sensitive to AK-induced nephrotoxicity and its uptake was reduced in a dose-dependent manner. This inhibition occurred at doses which did not alter BUN concentrations re-emphasizing the relative insensitivity of BUN as an indicator of AK-induced renal damage. One of the earliest events in the pathogenesis of AG-induced nephrotoxicity is the disruption of the lysosomal system with changes in lysosomal size, number and appearance 1221preceding other overt signs of tubular cell damage [23]. In both the present study and earlier reports by others [7,12,14], urinary NAG concentrations were elevated within l-3 days of the commencement of AG treatment. The early appearance of this lysosomal enzyme in the absence of brush border membrane damage is not surprising since, due to its solubility, NAG can be released from disrupted lysosomes across the brush border membrane into the tubular lumen [5]. In a study by Gibey et al. [12], it was proposed that a relationship between the
111
TABLE II CHANGES IN KIDNEY TUBULAR MORPHOLOGY TREATMENT WITH AMIKACIN (AK) Dose of AK
Control 100 (1) log (2) 200 (1) 200 (2) 400 (1) 400 (2) a The total daily bMorphological ’ Morphological d Mo~hological
Tubular dilation _b +c + + + ++d ++
IN GUINEA PIGS AFTER 10 DAYS OF
Tubular cell necrosis
Tubular regeneration
_
-
+ + -
+ + +
+ -
dose of AK was administered either in one injection (1) or two injections (2). changes absent. changes mild. changes moderate.
nephrotoxicity of GM and the initial urinary NAG activity may exist and that the degree of NAG response during the first 10 days of treatment may serve as a prognostic factor. It was further suggested by Donta and Lembke [I l] that the rate of increase to peak NAG excretion and the degree of toxicity could be correlated in patients receiving GM. In contrast to these studies, the results of the present study fail to show such relationships. Due to large intragroup variations, daily fluctuations and the absence of any distinct urinary excretion pattern for NAG, there does not appear to be any quantitative relationship between AK-induced renal damage and enzymuria. Large day-to-day variations in the urinary excretion of NAG in newborns were also reported by Adelman et al. [24] who concluded that because of these variations, NAG would not be particularly useful in predicting subsequent changes in renal function caused by AGs. In another clinical study by Fujita et al. 1251,it was also concluded that due to the large variation in the urinary NAG response in patients during AG therapy, the degree of nephrotoxicity of AG as a function of urinary NAG levels could not be shown. The sensitivity of the brush border membrane enzyme, ALP to AK-induced renal damage was also examined, however, similar to NAG, large daily variations in the urinary excretion of ALP were observed during treatment with AK in the one-injection groups. In contrast, with the two-injection regimen, a distinct peak in ALP excretion was observed. It was also observed that, for this dosing regimen, the higher the drug dose, the earlier the appearance of the peak in ALP excretion. In the 200 mg dose group (two injections/day regimen), the appearance of this peak (on day 9) approximately coincided with the decrease in TEA uptake which occurred by the eighth day of treatment. Kaloyanides 1261also reported sharp increases in enzymuria during the course of GM injection to rats which coincided with the onset of proximal
112
tubular cell necrosis. Thus, the peak urinary excretion of the brush border membrane enzyme, ALP in the two-injection group may reflect the onset of tubular cell necrosis. Why similar peaks in the excretions of NAG and ALP (in the one-injection regimen) with different doses of AK treatment did not occur is unknown but emphasizes the difficulty in utilizing enzymuria as a quantitative marker of AK-induced renal damage. The dosing schedule is known to be an important factor in the nephrotoxicity of AGs. Administration of GM in divided doses has been shown to result in greater damage to the kidney than the administration of the same dose of the drug in a single injection [15,27,28] presumably due to a greater renal accumulation of the drug following the multiple dosing [17,29-311. Because the renal uptake mechanism for AGs is a saturable process, the efficiency of uptake decreases as plasma concentrations increase [32]. Therefore, following multiple dosing, a larger proportion of the drug can be taken up into the kidney since plasma levels will be lower after each divided dose than those attained following administration of the drug in a single injection. In the present study, at the two lower doses, administration of AK in two injections per day resulted in higher drug concentrations in the kidney. There also appeared to be greater renal damage with this regimen since there was a greater reduction in TEA transport as compared to the one-injection regimen. In addition, at the dose of 200 mgikg body wt., the two-injection regimen also resulted in greater changes in tubular morphology than the one-injection regimen. The difference in toxicity between the two different dosing regimens could not be discerned from the enzymuria data again due to large intragroup variations, daity ~uctuations and an absence of discernable trends in the urinary excretion of NAG and ALP. The difference in toxicity between the two dosing regimens was not observed at the 100 mg dose levels. However, at this lower dose, morphological changes observed in both dosing regimen groups were mild and limited to tubular dilation. At the dose of 400 mgikg body wt., division of the dose did not result in either a greater accumulation of the drug in the kidney or in greater nephrotoxicity. It is likely that at this dose, even with the divided dose, saturation of the uptake mechanism had occurred and therefore renal accumulation of the drug was limited. In conclusion, results from this investigation into the nephrotoxicity of AK suggest that, despite significant increases in the urinary excretions of NAG and ALP following AK administration, these enzymes do not appear to be useful quantitative or prognostic markers of AG-induced renal damage due to the difficulty in establishing a relationship between enzymuria and kidney damage.
The authors wish to thank Messrs. C. Paul, Solomonraj, K. Kittle and L. Samure for their excellent technical assistance.
113 REFERENCES 1 Meyer, R.D. (1981) Amikacin. Ann. Intern. Med. 95,32&332. 2 Gingell, J.C. and Waterworth, P.M. (1968) Dose of gentamicin in patients with normal renal function and renal impairment. Br. Med. J. 2, 19-22. 3 Price, R.G. (1982) Urinary enzymes, nephrotoxicity and renal disease. Toxicology 23,99-134. 4 Palla, R., Marchitiello, M., Tuoni, M., Cirami, C., Giovannini, L., Bertelli, A.A.E. and Bertelli, A. (1985) Enzymuria in aminoglycoside-induced kidney damage. Comparative study of gentamicin, amikacin, sisomicin and netilmicin. Int. J. Clin. Pharmacol. Res. 5, 351-355. 5 Pipemo, E. (1981) Detection of drug induced nephrotoxicity with urinanalysis and enzymuria assessment. In: J.B. Hook (Ed.), Toxicology of the Kidney, Raven Press, New York, pp. 31-55. 6 Smith, J.H. and Hook, J.B. (1982) Experimental nephrotoxicity in vivo. In: P.H. Bach, R.W. Bonner, J.W. Bridges and E.A. Lock (Eds.), Nephrotoxicity Assessment and Pathogenesis, John Wiley and Sons, Chichester, pp. 117-127. 7 Garry, F., Chew, D.J. and Hoffsis, G.F. (1990) Enzymuria as an index of renal damage in sheep with induced aminoglycoside nephrotoxicosis. Am. J. Vet. Res. 51, 428432. 8 Luft, F.C., Pate], V., Yum, M.N., Patel, B. and Kleit, S.A. (1975) Experimental aminoglycoside nephrotoxicity. J. Lab. Clin. Med. 86, 213-220. 9 Patel, V., Luft, F.C., Yum, M.N., Patel, B., Zeeman, W. and Kleit, S.A. (1975) Enzymuria in gentamitin-induced kidney damage. Antimicrob. Agents Chemother. 8, 364369. 10 Wellwood, J.M., Lovell, D., Thompson, A& and Tighe, J.R. (1976) Renal damage caused by gentamicin: A study of the effects on renal morphology and urinary enzyme excretion. J. Pathol. 118, 171-182. 11 Donta, S.T. and Lembke, L.A. (1985) Com~rative effects of gentamicin and tobr~ycin on excretion of ~-acetyl-~-D-glucos~inida~. Antimicrob. Agents Chemother. 28, 500-503. 12 Gibey, R., Dupond, J.L., Alber, D., Lecorte des Floris, R. and Henry, J.C. (1981) predictive value of urinary N-acetyl-/I-n-glucosaminidase and p-2-microglobulin (82 M) in evaluating nephrotoxicity of gentamicin. Clin. Chim. Acta. 116,25-34. 13 Mondorf, A.W., Brier, J., Hindus, J., Scherberich, J.E., Mackenrodt, G., Shah, P.M., Stille, W. and Schoeppe, W. (1978) Effect of aminoglycosides on proximal tubular membranes of the human kidney. Eur. J. Clin. Pharmacol. 13, 133-142. 14 Palla, R., Paternoster, G., Galigani, P., Giovanni, L., Bertelli, A.A.E., Romano, M.R., Alessandri, M.G. and Bertelli, A. (1987) Comparative effects of gentamicin, amikacin and dactimicin on excretion of N-acetyl-/+glucosaminidase (NAG) and kidney histological pattern in rats. Drugs Exp. Clin. Res. 13,751-756. 15 Bennett, W.M., Plamp, C.E., Gilbert, D.N., Parker, R.A. and Porter, G.A. (1979) The influence of dosage regimen on ex~rimental gentamicin nephrotoxicity: Dissociation of peak serum levels from renal failure. J. Infect. Dis. 140, 576580. 16 Kluwe, W.M. and Hook, J.B. (1978) Functional nephrotoxicity of gentamicin in the rat. Toxicol. Appl. Pharmacol. 45,163-l 75. 17 Reiner, N.E., Bloxham, D.D. and Thompson, W.L. (1978) Nephrotoxicity of gentamicin and tobramytin given once daily or continuously in dogs. J. Antimicrob. Chemother. 4, 855101. 18 Giuliano, R.A., Verpooten, G.A., Pollet, DE., Verbist, L., Scharpe, S.L. and DeBroe, M.E. (1984) Improved procedure for extracting aminoglycosides from renal cortical tissue. Antimicrob. Agents Chemother. 25, 783-784. 19 Smith, J.H., Braselton, W.E., Tonsager, S.R., Mayor, G.H. and Hook, J.B. (1982) Effects of vanadate on organic ion accumulation in rat renal cortical slices. J. Pharmacol. Exp. Ther. 220, 540-546. 20 Carpenter, H.M. and Mudge, G.H. (1980) Uptake and acetylation of p-aminohippurate by slices of mouse kidney cortex. J. Pharmacol. Exp. Ther. 213,3X)-354. 21 Bennett, W-M., Plamp, C.E., Parker, R.A., Gilbert, D.N.. Houghton, D.C. and Porter, G.A. (1980)
114
Alterations 95, 32239.
in organic
22 Houghton,
ion transport
D.C., Hartnett,
electron
microscopic
23 Tulkens,
nisms and perspectives.
of gentamicin
K., Murayama,
26 Kaloyanides, tex. Fund. 27 Aronoff,
M., Porter,
nephrotoxicity
in the rat. J. Lab. Clin. Med.
G. and Bennett,
W. (1976) A light and
in rats. Am. J. Physiol.
studies on nephrotoxicity
of aminoglycosides
F. and Rubis, T. (1987) A controlled
in the neonate.
cosides on excretion
nephrotoxicity
82, 589-612. at low doses: Mecha-
Am. J. Med. 80, 5514.
R.D., Wirth,
and amikacin
by gentamicin
M., Campbell-Boswell,
analysis
P.M. (1986) Experimental
24 Adelman, 25 Fujita,
induced
T., Kawamura,
M. and Narita,
of N-acetyl-B-o-glucosaminidase.
G.R., Pottratz,
K. (1987) Comparison Int. J. Clin. Pharm.
G.J. (1984) Aminoglycoside-induced Appl. Toxicol.
study of the nephrotoxicity
of mezlocillin
Am. J. Dis. Child 141, 1175-l 178.
functional
of the effect of aminoglyRes. 7, 4777480.
and biochemical
defects in the renal cor-
4, 930-943. S.T., Brier, M.E., Walker,
(1983) Aminoglycosideaccumulation
N.E., Fineberg,
N.S., Glant,
kinetics in rat renal parenchyma.
M.D. and Luft, F.C.
Antimicrob.
Agents Chemother.
23, 74478. 28 Powell,
S.H.,
Groden,
D.L.,
Thompson, Jacobs,
W.L., M.R.,
continuous
aminoglycoside
netilmicin
and tobramycin.
Luthe,
Discenna,
dosing:
M.A., A.O.,
Efficacy
Stern, Cash,
and toxicity
R.C., H.A.
Grossniklaus,
and Klinger,
in animals
D.A.,
R.A., Verpooten,
kinetics of aminoglycosides 31 Verpooten, dosing 32 DeBroe,
G.A.,
decreases
Giuliano,
vs.
J. Infect. Dis. 147, 918-932.
G.A., Verbist,
L., Wedeen,
R.P. and DeBroe,
in the kidney cortex of rats. J. Pharmacol. R.A.,
renal accumulation
M.E., Giuliano,
D.D.,
and clinical studies of gentamicin,
29 Giulano, R.A., Verpooten, G.A. and DeBroe, M.E. (1986) The effect of dosing cortical accumulation of aminoglycosides in rats. Am. J. Kidney Dis. 8, 297-303. 30 Giuliano,
Bloxham,
J.D. (1983) Once daily
Verbist,
L., Eestermans,
of gentamicin
R.A. and Verpooten,
on kidney
M.E. (1986) In vivo uptake
Exp. Ther. 236, 470-475.
G. and DeBroe,
and netilmicin.
strategy
M.E. (1989) Once-daily
Clin. Pharmacol.
G.A. (1989) Aminoglycoside
Ther. 45, 22-27.
nephrotoxicity:
nism and prevention. In: A. Ameriao, P. Coratelli, V.M. Cambese and S.G. Massry Systemic Diseases and the Kidney, Plenum Publishing, New York, pp. 2333245.
Mecha-
(Eds), Drugs,