FUNDAMENTAL
AND
APPLIED
TOXICOLOGY
Nephrotoxicity MICHIYUIU
KATO,
MITSUYOSHI
18,532-539
(1992)
of a New Cephalosporin, YOSHIDA,
HARUMI
SHIMADA,
DQ-2556, in Rats
KAZUMI AKAHANE, AND SATOSHI TAKAYAMA
Drug Safety Research Center, Research Institute, Daiichi Pharmaceutical Co., Ltd. Received May 8, 199 1; accepted October 15, 199 1
Nephrotoxicity
of a New Cephalosporin, DQ-2556, in Rats.
KATO, M., YOSHIDA, TAKAYAMA,
M., SHIMADA,
H., AKAHANE,
K., AND
S. (1992). Fundam. Appl. Toxicol. 18, 532-539.
A single intravenous administration of a new cephalosporin, DQ-2556, at 1200 mg/kg to Sprague-Dawley rats induced proximal tubular necrosis. The histological lesions were not closely correlated with renal cortical concentrations of DQ-2556. Development of renal injuries was examined histologically. The arcuate and cortical radial arteries of the kidneys were constricted immediately after the injection, but then became dilated and showed histological changes: penetration by erythrocytes, edematous thickening, and necrosis in the media. In addition to congestion in the outer medulla, the proximal convoluted and straight tubules exhibited the earliest changes 30 min after the injection, namely, enlargement and rounding of the mitochondria and swelling and irregular arrangement of the microvilli in the epithelial cells. Small necrotic foci of epithelium were observed from 4 hr. They were mainly distributed in the outer cortex and the outer stripe of the outer medulla 24 hr later. In the examination of hemodynamics, DQ-2556 significantly decreased the blood flow with increased vascular resistance in the renal artery during and after a single injection. These changes had disappeared wholly or partially 60 min after dosing commenced. Furthermore, Ca*+ channel blockers markedly inhibited increases in the serum urea nitrogen and creatinine concentrations and development of the tubular necrosis and the lesions of the arterial walls, which were induced by DQ-2556. These results suggest a possible contribution of the constriction of renal artery to the tubular necrosis. o 1992 Society of Toxicology.
Renal ischemia and administration of nephrotoxic substances or potent diuretics are among the potential causes of renal injury (Lawson et al., 1972). The primary nephrotoxic antibiotics are aminoglycosides and cephalosporins (Bendirdjian et al., 1981). Among the cephalosporins, cephaloridine (CER) and cephaloglycin are reported to induce renal failure in both humans and animals. Therefore, their nephrotoxicity has been studied in detail using rabbits (Tune and Fravert, 1980a,b, Wold and Turnipseed, 1980) because this animal is more sensitive than other species to cephalosporin-induced renal failure (Atkinson et al., 1966). Rabbits are therefore recommended as appropriate models for as0272-0590/92
$5.00
Copyright 0 1992 by the Society of Toxicology. All rights of reproduction in any form reserved.
532
sessment of nephrotoxicity of cephalosporins (Silverblatt, 1982). DQ-2556 is a broad-spectrum cephalosporin synthesized by Ejima et al. (1987). This drug is structurally characterized by oxazole substitution in the pyridine group at the 3-position of the cephalosporin ring. DQ-2556 has excellent activity against clinically important gram-negative pathogens, and its antistaphylococcal activity is much better than that of ceftazidime. Furthermore, Haemophilus influenzae and Neisseria gonorrhoeae are also highly susceptible to the drug (Fujimoto et al., 1986). In laboratory animals, DQ-2556 has a different type of nephrotoxic potential from other cephalosporins, since a single injection of high dose causes necrosis of the proximal tubular epithelium of the kidney in rats, but not in rabbits. Additionally, we found lesions of the arterial walls associated with tubular necrosis in the kidneys of rats. Therefore, we have examined the time course of the histological and hemodynamic changes and discussed the relationship between them. METHODS Animals and preparation of drag. Male Sprague-Dawley rats aged 6-l weeks and weighing 300-470 g (for hemodynamic study) were used in the present study. The animals were purchased from Japan SLC, Inc. (Shizuoka) and housed 2-5 per cage in wire-mesh cages. They were maintained in airconditioned rooms (temperature, 23 + 2°C; humidity, 55 + 15%; light/dark cycle, 12/ 12 hr) and allowed free accessto commercial laboratory chow (F2, Funabashi Farm) and tap water. DQ-2556 was supplied by Daiichi Pharmaceutical Co., Ltd. and dissolved in distilled water at a concentration of 7.5% before administration. DQ-2556 concentration in renal cortex. A batch of 15 animals received a single bolus injection of DQ-2556 into the tail vein at a dose of 200,400, 800, or 1200 mg/kg, and they were killed by exsanguination under ether anesthesia 5 min, 6, and 24 hr after the administration. The cortex was resected from the kidney, weighed, and stored at -2O’C until measurement of the drug concentration by bioassay (thin-layer disc method) using Staphylococcus aureus 209P and Escherichia coli 3804C. Development of histological changes in the kidneys. Seventy animals were divided into control and treatment groups. The animals of the treatment group received a single bolus injection of DQ-2556 at a dose of 1200 ms/ kg. Batches of five animals from the treatment group were killed 0 (immediately), 30 min, and 1, 2, 4, 8, and 24 hr after the administration. The control animals received saline and were killed in batches at the same termination points as the treatment animals. A part of the kidney was removed under ether anesthesia and immersed in 3% glutaraldehyde-phosphate buffer solution. It was cut into small pieces,
NEPHROTOXICITY fixed in the same fixative, washed with the buffer solution, and postfixed in 1% osmium tetroxide-phosphate buffer solution. They were dehydrated with ethanol and embedded in epoxy resin (Quetol8 12). Ultrathin sections were made, stained with lead citrate and uranyl acetate, and examined under an electron microscope (H500, Hitachi). After the animal was terminated, the kidney was removed and fixed in buffered 10% formalin. Following dehydration with ethanol, they were embedded in pa&in wax, sectioned, stained with hematoxylin and eosin (H-E) and periodic acid-Schiff base, and examined with a light microscope. Histological scores were assigned in accordance with the following criteria: 0.5, necrosis of a few proximal tubular epithelial cells; 1, multiple foci of necrosis; 2, expanded necrosis in about 50% of the areas of the cortex and the outer stripe of the outer medulla; 3, necrosis in more than 50% of the areas. The luminal diameters of the arcuate and cortical radial arteries that appeared to have been cut perpendicularly to their long axes were simultaneously measured at the rate of lo-25 arteries per specimen with a micrometer (Olympus). Exumination of hemodynomics. Fifteen animals were used. After tracheotomy under anesthesia by inactin (i.p., 100 mg/kg), catheters were inserted and fixed in the femoral artery and vein, and an electromagnetic blood flow meter (probe diameter: 0.5 mm; Nihon Kohden) was placed around the renal artery. The other ends of the catheters and the blood flow meter were connected to a carrier amplifier (AP-60lG, Nihon Kohden), heart rate counter (ATdOOG), or DC amplifier (AD-601G). DQ-2556, 200 or 1200 mg/kg, was administered to five rats through the catheter into the vein at an injection speed of 1 ml/min, which was chosen to allow examination of detail changes of each parameter. The other five animals served as controls and received the same volume of saline as the high dose of DQ2556. Then, the arterial mean blood pressure, heart rate, and blood flow in the renal artery were recorded immediately before administration and 1, 3, 5, 10,20, 30, and 60 min after commencement of the administration. The vascular resistance of the renal artery was calculated from the values of mean blood pressure and blood flow. Efkcts of vasodilators on nephrotoxicity of DQ-2556. One hundred and eighty animals were used. Six vasodilatom having different actions (Table 1) were administered concurrently with or before the injection of DQ-2556, and their inhibitory effectson DQ-2556 nephrotoxicity were examined. Three dosage levels of each vasodilator were determined so as to comprise the clinical dose and two multiples of that dose. Verapamil, nicardipine, and dopamine were purchased from the pharmaceutical companies, Eisai, Yamanouchi, and Shionogi, respectively.They were diluted with saline to obtain
TABLE 1 Actions and Doses of Vasodilators Vasodilators Verapamil hydrochloride Nicardipine hydrochloride Dopamine hydrochloride Bamethan sulfate Trapidil Trimetazidine
hydrochloride
Action (route) doses Ca’+ channel blocker (i.v.) 0.0625, 0.125, or 0.25 &rat Car+ channel blocker (i.v.) 2.5,25, or 250 ccp/rat Dopaminergic agonist (i.v.) 12.5, 50, or 200 &rat Stimulation of @-receptor (p.0.) 1, 10, or 100 mg/kg Inhibition of thromboxane Az production (p.0.) 5, 20, or 80 mg./kg Relaxation of vasoconstriction induced by PGFz (p.0.) 0.1, 1, or 10 mg/kg
OF DQ-2556 IN RATS
533
the expected concentrations and perfused into the tail vein at 0.5 ml/min for 5 min using an infusion pump (Harvard), and a bolus injection of DQ2556, 1200 mg/kg, was performed 2.5 min after the infusion of the vasodilators commenced. Bamethan, trapidil, and trimetazidine were obtained from the pharmaceutical companies, Tanabe, Mochida, and Ohta, respectively. Because all of these drugs are in tablet form, they were powdered, suspended in 0.5% carboxymethylcellulose sodium solution, and administered orally at a constant volume of 10 ml/kg 1 or 2 hr before the injection of DQ-2556, at which times the drug concentration in the blood is expected to reach the maximum. Two days after the above treatments, blood was collected by transection of the cervical artery from the animals under ether anesthesia, and the kidneys were removed, weighed, and examined histologically. The serum was separated from each blood sample, and serum urea nitrogen (BUN) and creatinine (CRE) concentrations were determined using an autoanalyzer (Type 736, Hitachi). Statistical unulysis. For blood chemistry data, kidney weights, diameters of arteries, and hemodynamic data, Student’s t test was employed for statistical assessment of differences in the mean values between the control and treatment groups. RESULTS
DQ-2556 Concentration
in Renal Cortex
Concentration of DQ-2556 in renal cortex and individual comparison of the concentration with histological score are shown in Table 2. High values were obtained in all groups 5 min after injection of DQ-2556 although there was a discrepancy in the values between the 400 and 800 mg/kg groups. The values of the 200, 400, and 800 mg/kg groups were markedly decreased in a similar manner 24 hr later. However, the decrease was not sharp in the 1200 mg/kg group, and the value was kept higher than other groups 24 hr later. Histologically, proximal tubular necrosis was observed only in the 800 and 1200 mg/kg groups. There were large individual variations in the concentration of DQ-2556 and severity of histological changes. The 1200 mg/kg group showed more severe changes than the 800 mg/kg group, but individually, there was no clear relationship between the cortical concentrations of DQ2556 and appearance and severity of the histological changes. One rat showed moderate necrosis of tubules with the concentration of 11.2 pg/g cortex, but others showed mild and no lesion with 46.2 and 53.5 pug/g, respectively. Development
of Histological
Changes in the Kidneys
Time course of changes in the luminal diameters of the arteries in the kidneys after a single injection of DQ-2556 are presented in Fig. 1. The DQ-2556 treatment group showed a significant decrease in the diameter immediately after the injection, compared with the control value. The diameter then began to increase, and this increase became obvious at 2 hr and continued up to 24 hr later. Histological changes observed in the kidneys were as follows. Penetration by several erythrocytes in the media of a few arcuate or cortical radial arteries (or both) was observed 30
KATO
534
ET AL.
TABLE 2 Renal Cortical Concentrations of DQ-2556 and Individual Comparisons with Histological in Rats Killed after a Single Intravenous Injection
Scores
Renal cortical concentration of DQ-2556 Time after injection Grow bm/k)
5 min
200 400 800 1200
7,380.O 17,132.0 11,717.0 30,345.O
k + t +
6 hr 988.3" 2643.5 643.0 4130.8
86.8 24.5 69.5 5814.0
f + i iz
24
21.9 3.1 23.6 3311.0
4.3 3.6 8.8 291.9
hr + +t f
1.3 1.2 1.2 170.8
Individual comparison 800
Concentration of DQ-2556
no.
Rat
mg/kg
1
6.0* 10.2
2 3 4 5
1200
Histological score 0.5' 1.0
11.0
1.0
11.2 5.7
2.0 0.5
mg/kg
Rat no.
Concentration of DQ-2556
1
961.6
2 3 4 5
213.7
184.7 53.5 46.2
Histological score 3.0 2.0 2.0 0 1.0
a Mean f SE, @g/g cortex. * Microgram per gram cortex. ‘See the text.
min after a single injection of DQ-2556 (Fig. 2). Additionally, interstitial capillaries were congested in the outer medulla. From 1 hr later, epithelial cells of the proximal convoluted and straight tubules showed mild changes, which were vesiculation of cytoplasms and staining loss of the brush borders. Edematous swelling of apical cytoplasms and desquamation of the brush borders were seen 2 hr later (Fig. 3). Foci of proximal convoluted tubules with necrotic epithelium were
oh.?
,
0
0.5
2
, 4
1 8
4 24 hr
FIG. 1. Time course of changes in luminal diameters of intrarenal arteries in rats receiving a single administration of DQ-2556 (1200 mg/kg, i.v.). *, significant difference from the control value (p c: 0.05).
seen sporadically in the cortex 4 hr later. Necrotic areas of the tubules were expanded in course of time, and at 24 hr, they were distributed through the whole outer cortex, the whole outer stripe of the outer medulla, and patchy areas of the inner cortex in severe cases. Necrosis also was observed in the walls of dilated arteries. Electron-microscopic examination showed the medial smooth muscle cells having many cytoplasmic vacuoles in the arteries immediately (0 min) after injection of DQ-2556. At 30 min, the earliest changes were observed in the epithelial cells of the proximal tubules. They were swelling and irregular patterning of the microvilli and rounding and enlargement of the mitochondria. Thereafter, the epithelial cells showed a labyrinthine network of membrane profiles, densely packed vesicles, and whorls in the apical cytoplasms. The mitochondria were markedly enlarged and the electron density of the matrix was obviously decreased in some of them (Fig. 4). There also were changes which are thought to be reparative. The epithelium had well-developed microvilli in spite of marked cytoplasmic degeneration and other cells had many mitochondria of normal shape and size, but not microvilli. Examination
of Hemodynamics
Time course of hemodynamic changes in rats given DQ2556 are presented in Fig. 5. No parameters were signifi-
NEPHROTOXICITY
OF DQ-2556 IN RATS
535
FIG. 2. The arcuate artery in the kidney of a rat killed 30 min after a single administration of DQ-2556 (1200 mg/kg, i.v.). Penetration of erythrocytes into the media (arrows) is seen. H-E (X800).
cantly affected in the control and DQ-2556 (200 mg/kg) groups. Blood flow of the renal artery in the DQ-2556 (1200 mg/kg) group began to decline 5 min after dosing was commenced, reaching its lowest point of an approximately 50% decrease from the predosing value (mean f standard error, 4.4 f 0.3 ml/min). It gradually rose from 30 min, but had not completely recovered even 60 min later. The vascular resistance of the renal artery, on the other hand, was very markedly raised to about 2.1 times its predosing value (29 + 3.0 unit) from 10 to 20 min, although the difference was not statistically significant because of the large standard errors, and it had declined markedly by 60 min after the commencement of dosing. Mean blood pressure (123.7 & 5.0 mm Hg) was transiently elevated by about 20% during administration of the drug, but recovered its normal level 60 min later. Heart rate (401.2 f 16.6 beats/min) was gradually lowered by about 20% up to 60 min later.
Efects of Vasodilators on Nephrotoxicity
of DQ-2556
The results obtained from rats receiving 1200 mg/kg of DQ-2556 alone or in combination with vasodilators (three doses) are presented in Fig. 6. In all six experiments, injections of DQ-2556 markedly elevated the group mean values of BUN (2.6-3.8 times the control value) and of CRE (3.1-5.2 times), but the differences from the control group were not statistically significant in general because of large individual variations. DQ-2556 increased kidney weight and induced tubular necrosis (mean histological score, 1.7-2.0) with changes of the intrarenal arterial walls in all animals except for one case in the experiment for trapidil. The Ca2’ channel blockers verapamil and nicardipine markedly lowered the BUN and CRE values to the control levels and the histological scores to 0.5-0.9 at all three doses. Additionally, these drugs completely prevented vascular changes or decreased their incidence and severity. However, at these doses, there was no dose depen-
KATO ET AL.
FIG. 3. Proximal convoluted tubules in the kidney of a rat killed 2 hr after a single administration of DQ-2556 (1200 m&g, i.v.). Apical cytoplasms of epithelial cells are edematous and protruded into the lumens, and the cells have pycnotic nuclei and no brush borders. H-E (X425).
dency in the inhibitory effects of the vasodilators. Dopamine and trapidil showed an inhibition of the nephrotoxicity at low and high dosage levels, respectively. Bamethan and trimetazidine had no effect on the nephrotoxicity at all doses. Intravenous or oral administration of each vasodilator alone had no effect, and intravenous concurrent administration of DQ-2556 with saline instead of vasodilators showed similar effects on renal function and histology to those of DQ-2556 alone. DISCUSSION A single intravenous administration of 1200 mg/kg of DQ2556 induced marked reduction of blood flow with raised vascular resistance in the renal artery of rats during and after the injection. The luminal diameter of the arcuate and cortical radial arteries were decreased just aFter administration, but increased thereafter up to 24 hr later. Histologically, penetration of erythrocytes into the media was observed in
the dilated arteries, and medial necrosis ensued. The above findings suggest that DQ-2556 induces constriction of the renal artery and intrarenal arteries in rats. Both vasodilators and vasoconstrictors (Bhan et al., 1982; Giacomelli et al., 1976; Jones and Hurley, 1985; Joris et al., 1972; Kai et al., 1981; Kai-Yamamoto et al., 1982; Kerns et al., 1989;Kobori et al., 1979; Vitullo et al., 1985) have been reported to induce similar vascular lesions, which are hemorrhage and necrosis of the media in small or large caliber arteries including renal artery. The medial changes induced by DQ-2556 are also similar to those reported changes. Of various vasoactive compounds, angiotensin is thought to induce dilation of arteries with vascular lesions as an action secondary to the segmental constriction previously caused by it (Giese, 1964; Wiener and Giacomelli, 1973). Concerning vasoconstriction, the intracellular vacuoles caused by hemiation of one smooth muscle cell into another have been reported to be increased in number in arteries constricted by L-norepinephrine (Joris and Majno, 198 1). In accordance with this report, the present
NEPHROTOXICITY
OF DQ-2556 IN RATS
537
FIG. 4. Epithelial cells of a proximal convoluted tubule from the kidney of a rat killed 2 hr after a single administration of DQ-2556 (1200 mg/kg, i.v.). Mitochondria are more enlarged and have irregular shapes, and the electron density of the matrix of one of them is decreased. Lead citrate-uranyl acetate (X6760).
clear explanation in the literature about the relationship of spasm to medial injury of an artery. The present study has shown that a large dose of DQ-2556 has at least a constricting action on the renal and intrarenal arteries in rats. Histological examination revealed swelling of the microvilli and enlargement of the mitochondria as the earliest
study revealed many vacuoles in the medial smooth muscle cells of intrarenal arteries while they were constricted just after injection of DQ-2556 to rats. The arteries were subsequently dilated with leakage of erythrocytes into the media. However, it is unclear whether the leakage of erythrocytes preceded or followed the dilation of the arteries. There is no
Blood
flow
1L I 300
-
Con1r0l
.......’
DO-2556
200
---
DO-2556
1200
mglkg
Vascular
resistance
1
260.
mO/kg 220.
60 60 40
0,36
10 . . 20
30
---e----1 .._..........................
T 1
60 rni”
FIG. 5. Time course of changes in blood Ilow and vascular resistance in the renal artery of rats receiving a single administration of DQ-2556 (200 or 1200 mg/kg, i.v.). The curves are presented as percentage changes from predosing values (1009 o). * , significant difference from the control value (p < 0.05); #, significant difference from the predosing value (p < 0.05).
RAT0
538 Verapamil
Nicardipine
ET AL.
Dopamine
Bamethan
Trapidil
Trimetazidine
FIG. 6. Effects of vasodilators (three doses, see the text) on renal injuries induced by a single administration of DQ-2556 (1200 mg/kg, iv). BUN, blood urea nitrogen; CRE, creatinine; q , Control; n, DQ-2556; , DQ-2556 + vasodilator (low dose); Zl, DQ-2556 + vasodilator (middle dose); &I, DQ2556 + vasodilator (high dose); q l, vasodilator (high dose); *, significant difference from the control value (p < 0.05); #, significant difference from the DQ-2556 group (p < 0.05).
changes in the proximal tubular epithelium, from 30 min after the injection of DQ-2556. Desquamation of the brush borders and vesiculation of the cytoplasms were subsequently observed in the epithelial cells. Those findings have been included in the early changes of the kidney in rabbits, rats, and guinea pigs after administration of CER (Cojocel et al., 1983; Silverblatt et al., 1970; Silverblatt, 1982), and rabbits after cefotiam (Cojocel et al., 1988). Ischemia, too, has been reported to induce similar changes in the tubular epithelium of rats (Glaumann et al., 1975; Glaumann and Trump, 1975; Reimer et al., 1972; Venkatachalam et al., 1978). Thus, renal histological changes induced by cephalosporins are similar to ischemic changes. DQ-2556-induced changes are also similar to them and are suspected to be correlated to the constriction of the renal artery, being associated with early constriction of the intrarenal arteries. CER has been reported to inhibit the respiration of the mitochondria of the tubular epithelium (Tune and Hsu, 1990), and the inhibition has been shown to precede the earliest histological changes (Tune and Fravert, 1980a,b). Therefore, Tune and Fravert (1980b) consider that the inhibition of mitochondrial respiration is the most important molecular event in CER-induced renal injury. Respiratory inhibition of the mitochondria has also been demonstrated in ischemic renal injuries in rabbits (Tune and Hsu, 1990). Increased membrane permeability of the mitochondria, caused by peroxidation following to an increase in free radicals, has been reported to reduce production of ATP and to induce cell injury (Mason, 1986). On the basis of the above, it is possible that the key injury induced by DQ-2556, which
led to necrosis of the proximal tubular epithelium, is swelling of the mitochondria, which was seen in the early stages of renal injury. The Ca2+ channel blockers verapamil and nicardipine markedly, but not completely, inhibited tubular necrosis induced by DQ-2556. Verapamil and another Ca2+ channel blocker, nifedipine, have been reported to improve renal failure caused by contrast media, gentamicin or cyclosporine A (CSA) and to prevent renovasoconstriction induced by CSA (Rossi et al., 1989), norepinephrine, angiotensin, and vasopressin (Goldberg and Schrier, 1984). Respiratory inhibition of the mitochondria with the accumulation of Ca2’ has been reported in renal injury after norepinephrine-induced ischemia, and the inhibition is thought to be prevented by the above two vasodilators through retardation of the Ca*+ influx into the renal vascular cells and the tubular epithelium (Burke et al., 1984). From the above descriptions, it would be easy to attribute DQ-2556-induced-constriction of renal arteries and -necrosis of tubular epithelia to an increased influx of Ca*+ into the arterial smooth muscle cells and the tubular epithelial cells, and inhibition of renal injury to reduction of the influx by the Ca2+ channel blockers. Those mechanisms are thought to work in a large part for the DQ2556 nephrotoxicity in rats, but effects of other vasoconstrictive substances should be taken in consideration. In conclusion, DQ-2556-induced necrosis of the proximal tubular epithelial cells in the kidneys of rats is in a different manner from that of other cephalosporins generally showing close correlation between tubular necrosis and drug concentration in the renal cortex (Tune and Fravert, 1980a). DQ-
NEPHROTOXICITY
2556 simultaneously produced histological changes in the intrarenal arterial walls and reduced blood flow in the renal artery. Ca2+ channel blockers markedly inhibited both the tubular and vascular lesions, suggesting a correlation between tubular necrosis and vasoconstriction. ACKNOWLEDGMENTS The authors thank Dr. Shinjiro Hashimoto and Ms. Kumi Yoshida for measurement of drug concentrations in the kidneys.
REFERENCES Atkinson, R. M., Currie, J. P., Davis, B., Pratt, D. A. H., Sharpe, H. M.. and Tomich, E. G. (1966). Acute toxicity of cephaloridine, an antibiotic derived from cephalosporin C. Toxicol. Appi. Pharmacol. 8, 398-406. Bendirdjian, J.-P., Prime, D. J., Browning, M. C., Hsu, C.-Y., and Tune, B. M. (198 1). Additive nephrotoxicity of cephalosporins and aminoglycosides in the rabbit. J. Pharmacol. Exp. Thu. 218, 68 l-685. Bhan, R. D., Giacomelli, F., and Weiner, J. (1982). Adrenoceptor blockade in angiotensin-induced hypertension. Am. J. Pathol. 108, 60-7 1. Burke, T. J., Arnold, P. E., Gordon, J. A., Bulger, R. E., Dobyan, D. D., and Schrier, R. W. (1984). Protective effect of intrarenal calcium membrane blockers before or after renal ischemia. Functional, morphological. and mitochondrial studies. J. Clin. Invest. 74, 1830- 184 1. Cojocel, C., Smith, J. H., Maita, K., Sleight, S. D., and Hook, J. B. (1983). Renal protein degradation: A biochemical target of specific nephrotoxicants. Fundam. Appl. Toxicol. 3, 278-284. Cojocel, C., Gottsche, U., Tolle, K.-L., and Baumann, K. (1988). Nephrotoxic potential of first-, second-, and third-generation cephalosporins. Arch. Toxicol. 62, 458-464. Ejima. A., Hayano, T., Ebata, T., Nagahara, T.. Koda, H., Tagawa, H., and Furukawa. M. (1987). Synthesis and antimicrobial activity of cephalosporins with a 1-pyridinium substituent carrying a 5-membered heterccycle at the C-3 position. J. Antibiot. 40, 43-48. Fujimoto, T.. Otani. T., Nakajima, R., Une, T., and Osada, Y. (1986). In vitro activity of DQ-2556, a new cephalosporin. Antimicrob. Agents Chemother. 30, 6 1 l-6 13. Giacomelli. F., Anversa, P.. and Wiener, J. (1976). Effect of angiotensininduced hypertension on rat coronary arteries and myocardium. .4m. J. Pathol. 84, 11 I-138. Giese. J. (1964). Acute hypertensive vascular disease. 2. Studies on vascular reaction patterns and permeability changes by means of vital microscopy and colloidal tracer technique. Acta Pathol. Microbial. Stand. 62, 497515. Glaumann. B.. Glaumann. H., Berezesky, I. K., and Trump, B. F. (1975). Studies on the pathogenesis of ischemic cell injury. Il. Morphological changes of the pars convoluta (P, and P2) of the proximal tubule of the rat kidney made ischemic in vivo. Virchows Arch. B Cell Pathol. 19,28 l302. Glaumann, B., and Trump, B. F. (1975). Studies on the pathogenesis of ischemic cell injury. Ill. Morphological changes ofthe proximal pars recta tubules (P,) of the rat kidney made ischemic in vivo. Virchows Arch. B Cell Pathol. 19, 303-323. Goldberg, J. P., and Schrier, R. W. (1984). Effect of calcium membrane blockers on in vivo vasoconstrictor properties of norepinephrine, angiotensin II and vasopressin. Miner E/e&o&e Metab. 10, 178-I 83. Jones. G. M., and Hurley, J. V. (1985). The response of small blood vessels in rat skin and skeletal muscle to repeated application of histamine: A
OF DQ-2556 IN RATS
539
chemical. topographical, and electron microscopic study. J. Pathol. 146, 291-299. Joris. I., Majno, G.. and Ryan, G. B. (1972). Endothelial contraction in vivo: A study of the rat mesentery. Virchows Arch. B Cell Pathol. 12,7383. Joris. I.. and Majno, G. (198 1). Medial changes in arterial spasm induced by L-norepinephrine. Am. J. Pathol. 105, 2 12-222. Kai, M., Kanaide, H., Yamamoto, H., Kurozumi. T., Tanaka, K., and Nakamura, M. (198 1). The pathogenesis of small arterial lesions in nephrectomized rats by the administration of renin. Br. J. Exp. Pathol. 62, 136141. Kai-Yamamoto, M., Kanaide, H., Yamamoto, H., Kurozumi, T., Kato, K., and Nakamura, M. (1982). The production of small arterial lesions in nephrectomized rats by the administration of pure cathepsin D. Br. J. Exp. Pathol. 63, 501-505. Kerns, W. D., Arena, E., and Morgan, D. G. (1989). Role of dopaminergic and adrenergic receptors in the pathogenesis of arterial lesions induced by fenoldopam mesylate and dopamine in the rat. Am. J. Pathol. 135, 339-349. Kobori, K., Suzuki, K.. Yoshida, Y., and Ooneda, G. (1979). Light and electron microscopic studies on rat arterial lesions induced by experimental arterial contraction. Virchows Arch. A Pathol. Anat. Histol. 385, 29-39. Lawson, D. H., Macadam. R. F., Singh, H., Gavras, H., Hartz, S., Turnbull, D., and Linton, A. L. (1972). Effect of furosemide on antibiotic-induced renal damage in rats. J. Infect. Dis. 126, 593-600. Mason, J. (1986). The pathophysiology of ischemic acute renal failure. A new hypothesis about the initiation phase. Renal Physiol. 9, 129- 147. Reimer, K. A., Ganote, C. E., and Jennings. R. B. (1972). Alterations in renal cortex following ischemic injury. Ill. Ultrastructure of proximal tubules after ischemia or autolysis. Lab. Invest. 26, 347-363. Rossi, N. F., Churchill, P. C., Mcdonald, F. D., and Ellis, V. R. (1989). Mechanism of cyclosporine A-induced renal vasoconstriction in the rat. J. Pharmacol. Exp. Ther. 250, 896-901. Silverblatt, F., Turck, M., and Bulger, R. (1970). Nephrotoxicity due to cephaloridine: A light- and electron-microscopic study in rabbits. J. Infect. Dis. 122, 33-44. Silverblatt, F. (1982). Pathogenesis of nephrotoxicity of cephalosporins and aminoglycosides: A review of current concepts. Rev. Infect. Dis. 4(Suppl.), S360-S365. Tune, B. M., and Fravert, D. (1980a). Mechanisms of cephalosporin nephrotoxicity: A comparison of cephaloridine and cephaloglycin. Kidney Int. l&591-600.
Tune, B. M., and Fravert, D. (1980b). Cephalosporin nephrotoxicity. Transport. cytotoxicity and mitochondrial toxicity of cephaloglycin. J. Pharmacol. Exp. Ther. 215, 186-190. Tune, B. M.. and Hsu, C.-Y. (1990). The renal mitochondrial toxicity of cephalosporins: Specificity of the effect on anionic substrate uptake. J. Pharmacol. Exp. Ther. 252,65-67. Venkatachalam, M. A., Bernard, D. B., Donohoe, J. F.. and Levinsky, N. G. (1978). lschemic damage and repair in the rat proximal tubule: Differences among the S, , SZ, and S3 segments. Kidney Int. 14, 3 l-49. Vitullo, J. C.. Genity, R. G., and Khairallah. P. A. (1985). Cellular changes in mesenteric arteries and veins after acute perfusions of angiotensin II and vasoactive amines. Blood Vessel 22, 286-300. Wiener, J., and Giacomelli, F. (1973). The cellular pathology of experimental hypertension. VII. Structure and permeability ofthe mesenteric vasculature in angiotensin-induced hypertension. Am. J. Pathol. 72, 22 I-240. Wold, J. S.. and Tumipseed, S. A. (1980). The effect of renal cation transport inhibitors on the in vivo and in vitro accumulation and efflux of cephaloridine. f-i& Sci. 27, 2559-2564.
AND
APPLIED
TOXICOLOGY
Nephrotoxicity MICHIYUIU
KATO,
MITSUYOSHI
18,532-539
(1992)
of a New Cephalosporin, YOSHIDA,
HARUMI
SHIMADA,
DQ-2556, in Rats
KAZUMI AKAHANE, AND SATOSHI TAKAYAMA
Drug Safety Research Center, Research Institute, Daiichi Pharmaceutical Co., Ltd. Received May 8, 199 1; accepted October 15, 199 1
Nephrotoxicity
of a New Cephalosporin, DQ-2556, in Rats.
KATO, M., YOSHIDA, TAKAYAMA,
M., SHIMADA,
H., AKAHANE,
K., AND
S. (1992). Fundam. Appl. Toxicol. 18, 532-539.
A single intravenous administration of a new cephalosporin, DQ-2556, at 1200 mg/kg to Sprague-Dawley rats induced proximal tubular necrosis. The histological lesions were not closely correlated with renal cortical concentrations of DQ-2556. Development of renal injuries was examined histologically. The arcuate and cortical radial arteries of the kidneys were constricted immediately after the injection, but then became dilated and showed histological changes: penetration by erythrocytes, edematous thickening, and necrosis in the media. In addition to congestion in the outer medulla, the proximal convoluted and straight tubules exhibited the earliest changes 30 min after the injection, namely, enlargement and rounding of the mitochondria and swelling and irregular arrangement of the microvilli in the epithelial cells. Small necrotic foci of epithelium were observed from 4 hr. They were mainly distributed in the outer cortex and the outer stripe of the outer medulla 24 hr later. In the examination of hemodynamics, DQ-2556 significantly decreased the blood flow with increased vascular resistance in the renal artery during and after a single injection. These changes had disappeared wholly or partially 60 min after dosing commenced. Furthermore, Ca*+ channel blockers markedly inhibited increases in the serum urea nitrogen and creatinine concentrations and development of the tubular necrosis and the lesions of the arterial walls, which were induced by DQ-2556. These results suggest a possible contribution of the constriction of renal artery to the tubular necrosis. o 1992 Society of Toxicology.
Renal ischemia and administration of nephrotoxic substances or potent diuretics are among the potential causes of renal injury (Lawson et al., 1972). The primary nephrotoxic antibiotics are aminoglycosides and cephalosporins (Bendirdjian et al., 1981). Among the cephalosporins, cephaloridine (CER) and cephaloglycin are reported to induce renal failure in both humans and animals. Therefore, their nephrotoxicity has been studied in detail using rabbits (Tune and Fravert, 1980a,b, Wold and Turnipseed, 1980) because this animal is more sensitive than other species to cephalosporin-induced renal failure (Atkinson et al., 1966). Rabbits are therefore recommended as appropriate models for as0272-0590/92
$5.00
Copyright 0 1992 by the Society of Toxicology. All rights of reproduction in any form reserved.
532
sessment of nephrotoxicity of cephalosporins (Silverblatt, 1982). DQ-2556 is a broad-spectrum cephalosporin synthesized by Ejima et al. (1987). This drug is structurally characterized by oxazole substitution in the pyridine group at the 3-position of the cephalosporin ring. DQ-2556 has excellent activity against clinically important gram-negative pathogens, and its antistaphylococcal activity is much better than that of ceftazidime. Furthermore, Haemophilus influenzae and Neisseria gonorrhoeae are also highly susceptible to the drug (Fujimoto et al., 1986). In laboratory animals, DQ-2556 has a different type of nephrotoxic potential from other cephalosporins, since a single injection of high dose causes necrosis of the proximal tubular epithelium of the kidney in rats, but not in rabbits. Additionally, we found lesions of the arterial walls associated with tubular necrosis in the kidneys of rats. Therefore, we have examined the time course of the histological and hemodynamic changes and discussed the relationship between them. METHODS Animals and preparation of drag. Male Sprague-Dawley rats aged 6-l weeks and weighing 300-470 g (for hemodynamic study) were used in the present study. The animals were purchased from Japan SLC, Inc. (Shizuoka) and housed 2-5 per cage in wire-mesh cages. They were maintained in airconditioned rooms (temperature, 23 + 2°C; humidity, 55 + 15%; light/dark cycle, 12/ 12 hr) and allowed free accessto commercial laboratory chow (F2, Funabashi Farm) and tap water. DQ-2556 was supplied by Daiichi Pharmaceutical Co., Ltd. and dissolved in distilled water at a concentration of 7.5% before administration. DQ-2556 concentration in renal cortex. A batch of 15 animals received a single bolus injection of DQ-2556 into the tail vein at a dose of 200,400, 800, or 1200 mg/kg, and they were killed by exsanguination under ether anesthesia 5 min, 6, and 24 hr after the administration. The cortex was resected from the kidney, weighed, and stored at -2O’C until measurement of the drug concentration by bioassay (thin-layer disc method) using Staphylococcus aureus 209P and Escherichia coli 3804C. Development of histological changes in the kidneys. Seventy animals were divided into control and treatment groups. The animals of the treatment group received a single bolus injection of DQ-2556 at a dose of 1200 ms/ kg. Batches of five animals from the treatment group were killed 0 (immediately), 30 min, and 1, 2, 4, 8, and 24 hr after the administration. The control animals received saline and were killed in batches at the same termination points as the treatment animals. A part of the kidney was removed under ether anesthesia and immersed in 3% glutaraldehyde-phosphate buffer solution. It was cut into small pieces,
NEPHROTOXICITY fixed in the same fixative, washed with the buffer solution, and postfixed in 1% osmium tetroxide-phosphate buffer solution. They were dehydrated with ethanol and embedded in epoxy resin (Quetol8 12). Ultrathin sections were made, stained with lead citrate and uranyl acetate, and examined under an electron microscope (H500, Hitachi). After the animal was terminated, the kidney was removed and fixed in buffered 10% formalin. Following dehydration with ethanol, they were embedded in pa&in wax, sectioned, stained with hematoxylin and eosin (H-E) and periodic acid-Schiff base, and examined with a light microscope. Histological scores were assigned in accordance with the following criteria: 0.5, necrosis of a few proximal tubular epithelial cells; 1, multiple foci of necrosis; 2, expanded necrosis in about 50% of the areas of the cortex and the outer stripe of the outer medulla; 3, necrosis in more than 50% of the areas. The luminal diameters of the arcuate and cortical radial arteries that appeared to have been cut perpendicularly to their long axes were simultaneously measured at the rate of lo-25 arteries per specimen with a micrometer (Olympus). Exumination of hemodynomics. Fifteen animals were used. After tracheotomy under anesthesia by inactin (i.p., 100 mg/kg), catheters were inserted and fixed in the femoral artery and vein, and an electromagnetic blood flow meter (probe diameter: 0.5 mm; Nihon Kohden) was placed around the renal artery. The other ends of the catheters and the blood flow meter were connected to a carrier amplifier (AP-60lG, Nihon Kohden), heart rate counter (ATdOOG), or DC amplifier (AD-601G). DQ-2556, 200 or 1200 mg/kg, was administered to five rats through the catheter into the vein at an injection speed of 1 ml/min, which was chosen to allow examination of detail changes of each parameter. The other five animals served as controls and received the same volume of saline as the high dose of DQ2556. Then, the arterial mean blood pressure, heart rate, and blood flow in the renal artery were recorded immediately before administration and 1, 3, 5, 10,20, 30, and 60 min after commencement of the administration. The vascular resistance of the renal artery was calculated from the values of mean blood pressure and blood flow. Efkcts of vasodilators on nephrotoxicity of DQ-2556. One hundred and eighty animals were used. Six vasodilatom having different actions (Table 1) were administered concurrently with or before the injection of DQ-2556, and their inhibitory effectson DQ-2556 nephrotoxicity were examined. Three dosage levels of each vasodilator were determined so as to comprise the clinical dose and two multiples of that dose. Verapamil, nicardipine, and dopamine were purchased from the pharmaceutical companies, Eisai, Yamanouchi, and Shionogi, respectively.They were diluted with saline to obtain
TABLE 1 Actions and Doses of Vasodilators Vasodilators Verapamil hydrochloride Nicardipine hydrochloride Dopamine hydrochloride Bamethan sulfate Trapidil Trimetazidine
hydrochloride
Action (route) doses Ca’+ channel blocker (i.v.) 0.0625, 0.125, or 0.25 &rat Car+ channel blocker (i.v.) 2.5,25, or 250 ccp/rat Dopaminergic agonist (i.v.) 12.5, 50, or 200 &rat Stimulation of @-receptor (p.0.) 1, 10, or 100 mg/kg Inhibition of thromboxane Az production (p.0.) 5, 20, or 80 mg./kg Relaxation of vasoconstriction induced by PGFz (p.0.) 0.1, 1, or 10 mg/kg
OF DQ-2556 IN RATS
533
the expected concentrations and perfused into the tail vein at 0.5 ml/min for 5 min using an infusion pump (Harvard), and a bolus injection of DQ2556, 1200 mg/kg, was performed 2.5 min after the infusion of the vasodilators commenced. Bamethan, trapidil, and trimetazidine were obtained from the pharmaceutical companies, Tanabe, Mochida, and Ohta, respectively. Because all of these drugs are in tablet form, they were powdered, suspended in 0.5% carboxymethylcellulose sodium solution, and administered orally at a constant volume of 10 ml/kg 1 or 2 hr before the injection of DQ-2556, at which times the drug concentration in the blood is expected to reach the maximum. Two days after the above treatments, blood was collected by transection of the cervical artery from the animals under ether anesthesia, and the kidneys were removed, weighed, and examined histologically. The serum was separated from each blood sample, and serum urea nitrogen (BUN) and creatinine (CRE) concentrations were determined using an autoanalyzer (Type 736, Hitachi). Statistical unulysis. For blood chemistry data, kidney weights, diameters of arteries, and hemodynamic data, Student’s t test was employed for statistical assessment of differences in the mean values between the control and treatment groups. RESULTS
DQ-2556 Concentration
in Renal Cortex
Concentration of DQ-2556 in renal cortex and individual comparison of the concentration with histological score are shown in Table 2. High values were obtained in all groups 5 min after injection of DQ-2556 although there was a discrepancy in the values between the 400 and 800 mg/kg groups. The values of the 200, 400, and 800 mg/kg groups were markedly decreased in a similar manner 24 hr later. However, the decrease was not sharp in the 1200 mg/kg group, and the value was kept higher than other groups 24 hr later. Histologically, proximal tubular necrosis was observed only in the 800 and 1200 mg/kg groups. There were large individual variations in the concentration of DQ-2556 and severity of histological changes. The 1200 mg/kg group showed more severe changes than the 800 mg/kg group, but individually, there was no clear relationship between the cortical concentrations of DQ2556 and appearance and severity of the histological changes. One rat showed moderate necrosis of tubules with the concentration of 11.2 pg/g cortex, but others showed mild and no lesion with 46.2 and 53.5 pug/g, respectively. Development
of Histological
Changes in the Kidneys
Time course of changes in the luminal diameters of the arteries in the kidneys after a single injection of DQ-2556 are presented in Fig. 1. The DQ-2556 treatment group showed a significant decrease in the diameter immediately after the injection, compared with the control value. The diameter then began to increase, and this increase became obvious at 2 hr and continued up to 24 hr later. Histological changes observed in the kidneys were as follows. Penetration by several erythrocytes in the media of a few arcuate or cortical radial arteries (or both) was observed 30
KATO
534
ET AL.
TABLE 2 Renal Cortical Concentrations of DQ-2556 and Individual Comparisons with Histological in Rats Killed after a Single Intravenous Injection
Scores
Renal cortical concentration of DQ-2556 Time after injection Grow bm/k)
5 min
200 400 800 1200
7,380.O 17,132.0 11,717.0 30,345.O
k + t +
6 hr 988.3" 2643.5 643.0 4130.8
86.8 24.5 69.5 5814.0
f + i iz
24
21.9 3.1 23.6 3311.0
4.3 3.6 8.8 291.9
hr + +t f
1.3 1.2 1.2 170.8
Individual comparison 800
Concentration of DQ-2556
no.
Rat
mg/kg
1
6.0* 10.2
2 3 4 5
1200
Histological score 0.5' 1.0
11.0
1.0
11.2 5.7
2.0 0.5
mg/kg
Rat no.
Concentration of DQ-2556
1
961.6
2 3 4 5
213.7
184.7 53.5 46.2
Histological score 3.0 2.0 2.0 0 1.0
a Mean f SE, @g/g cortex. * Microgram per gram cortex. ‘See the text.
min after a single injection of DQ-2556 (Fig. 2). Additionally, interstitial capillaries were congested in the outer medulla. From 1 hr later, epithelial cells of the proximal convoluted and straight tubules showed mild changes, which were vesiculation of cytoplasms and staining loss of the brush borders. Edematous swelling of apical cytoplasms and desquamation of the brush borders were seen 2 hr later (Fig. 3). Foci of proximal convoluted tubules with necrotic epithelium were
oh.?
,
0
0.5
2
, 4
1 8
4 24 hr
FIG. 1. Time course of changes in luminal diameters of intrarenal arteries in rats receiving a single administration of DQ-2556 (1200 mg/kg, i.v.). *, significant difference from the control value (p c: 0.05).
seen sporadically in the cortex 4 hr later. Necrotic areas of the tubules were expanded in course of time, and at 24 hr, they were distributed through the whole outer cortex, the whole outer stripe of the outer medulla, and patchy areas of the inner cortex in severe cases. Necrosis also was observed in the walls of dilated arteries. Electron-microscopic examination showed the medial smooth muscle cells having many cytoplasmic vacuoles in the arteries immediately (0 min) after injection of DQ-2556. At 30 min, the earliest changes were observed in the epithelial cells of the proximal tubules. They were swelling and irregular patterning of the microvilli and rounding and enlargement of the mitochondria. Thereafter, the epithelial cells showed a labyrinthine network of membrane profiles, densely packed vesicles, and whorls in the apical cytoplasms. The mitochondria were markedly enlarged and the electron density of the matrix was obviously decreased in some of them (Fig. 4). There also were changes which are thought to be reparative. The epithelium had well-developed microvilli in spite of marked cytoplasmic degeneration and other cells had many mitochondria of normal shape and size, but not microvilli. Examination
of Hemodynamics
Time course of hemodynamic changes in rats given DQ2556 are presented in Fig. 5. No parameters were signifi-
NEPHROTOXICITY
OF DQ-2556 IN RATS
535
FIG. 2. The arcuate artery in the kidney of a rat killed 30 min after a single administration of DQ-2556 (1200 mg/kg, i.v.). Penetration of erythrocytes into the media (arrows) is seen. H-E (X800).
cantly affected in the control and DQ-2556 (200 mg/kg) groups. Blood flow of the renal artery in the DQ-2556 (1200 mg/kg) group began to decline 5 min after dosing was commenced, reaching its lowest point of an approximately 50% decrease from the predosing value (mean f standard error, 4.4 f 0.3 ml/min). It gradually rose from 30 min, but had not completely recovered even 60 min later. The vascular resistance of the renal artery, on the other hand, was very markedly raised to about 2.1 times its predosing value (29 + 3.0 unit) from 10 to 20 min, although the difference was not statistically significant because of the large standard errors, and it had declined markedly by 60 min after the commencement of dosing. Mean blood pressure (123.7 & 5.0 mm Hg) was transiently elevated by about 20% during administration of the drug, but recovered its normal level 60 min later. Heart rate (401.2 f 16.6 beats/min) was gradually lowered by about 20% up to 60 min later.
Efects of Vasodilators on Nephrotoxicity
of DQ-2556
The results obtained from rats receiving 1200 mg/kg of DQ-2556 alone or in combination with vasodilators (three doses) are presented in Fig. 6. In all six experiments, injections of DQ-2556 markedly elevated the group mean values of BUN (2.6-3.8 times the control value) and of CRE (3.1-5.2 times), but the differences from the control group were not statistically significant in general because of large individual variations. DQ-2556 increased kidney weight and induced tubular necrosis (mean histological score, 1.7-2.0) with changes of the intrarenal arterial walls in all animals except for one case in the experiment for trapidil. The Ca2’ channel blockers verapamil and nicardipine markedly lowered the BUN and CRE values to the control levels and the histological scores to 0.5-0.9 at all three doses. Additionally, these drugs completely prevented vascular changes or decreased their incidence and severity. However, at these doses, there was no dose depen-
KATO ET AL.
FIG. 3. Proximal convoluted tubules in the kidney of a rat killed 2 hr after a single administration of DQ-2556 (1200 m&g, i.v.). Apical cytoplasms of epithelial cells are edematous and protruded into the lumens, and the cells have pycnotic nuclei and no brush borders. H-E (X425).
dency in the inhibitory effects of the vasodilators. Dopamine and trapidil showed an inhibition of the nephrotoxicity at low and high dosage levels, respectively. Bamethan and trimetazidine had no effect on the nephrotoxicity at all doses. Intravenous or oral administration of each vasodilator alone had no effect, and intravenous concurrent administration of DQ-2556 with saline instead of vasodilators showed similar effects on renal function and histology to those of DQ-2556 alone. DISCUSSION A single intravenous administration of 1200 mg/kg of DQ2556 induced marked reduction of blood flow with raised vascular resistance in the renal artery of rats during and after the injection. The luminal diameter of the arcuate and cortical radial arteries were decreased just aFter administration, but increased thereafter up to 24 hr later. Histologically, penetration of erythrocytes into the media was observed in
the dilated arteries, and medial necrosis ensued. The above findings suggest that DQ-2556 induces constriction of the renal artery and intrarenal arteries in rats. Both vasodilators and vasoconstrictors (Bhan et al., 1982; Giacomelli et al., 1976; Jones and Hurley, 1985; Joris et al., 1972; Kai et al., 1981; Kai-Yamamoto et al., 1982; Kerns et al., 1989;Kobori et al., 1979; Vitullo et al., 1985) have been reported to induce similar vascular lesions, which are hemorrhage and necrosis of the media in small or large caliber arteries including renal artery. The medial changes induced by DQ-2556 are also similar to those reported changes. Of various vasoactive compounds, angiotensin is thought to induce dilation of arteries with vascular lesions as an action secondary to the segmental constriction previously caused by it (Giese, 1964; Wiener and Giacomelli, 1973). Concerning vasoconstriction, the intracellular vacuoles caused by hemiation of one smooth muscle cell into another have been reported to be increased in number in arteries constricted by L-norepinephrine (Joris and Majno, 198 1). In accordance with this report, the present
NEPHROTOXICITY
OF DQ-2556 IN RATS
537
FIG. 4. Epithelial cells of a proximal convoluted tubule from the kidney of a rat killed 2 hr after a single administration of DQ-2556 (1200 mg/kg, i.v.). Mitochondria are more enlarged and have irregular shapes, and the electron density of the matrix of one of them is decreased. Lead citrate-uranyl acetate (X6760).
clear explanation in the literature about the relationship of spasm to medial injury of an artery. The present study has shown that a large dose of DQ-2556 has at least a constricting action on the renal and intrarenal arteries in rats. Histological examination revealed swelling of the microvilli and enlargement of the mitochondria as the earliest
study revealed many vacuoles in the medial smooth muscle cells of intrarenal arteries while they were constricted just after injection of DQ-2556 to rats. The arteries were subsequently dilated with leakage of erythrocytes into the media. However, it is unclear whether the leakage of erythrocytes preceded or followed the dilation of the arteries. There is no
Blood
flow
1L I 300
-
Con1r0l
.......’
DO-2556
200
---
DO-2556
1200
mglkg
Vascular
resistance
1
260.
mO/kg 220.
60 60 40
0,36
10 . . 20
30
---e----1 .._..........................
T 1
60 rni”
FIG. 5. Time course of changes in blood Ilow and vascular resistance in the renal artery of rats receiving a single administration of DQ-2556 (200 or 1200 mg/kg, i.v.). The curves are presented as percentage changes from predosing values (1009 o). * , significant difference from the control value (p < 0.05); #, significant difference from the predosing value (p < 0.05).
RAT0
538 Verapamil
Nicardipine
ET AL.
Dopamine
Bamethan
Trapidil
Trimetazidine
FIG. 6. Effects of vasodilators (three doses, see the text) on renal injuries induced by a single administration of DQ-2556 (1200 mg/kg, iv). BUN, blood urea nitrogen; CRE, creatinine; q , Control; n, DQ-2556; , DQ-2556 + vasodilator (low dose); Zl, DQ-2556 + vasodilator (middle dose); &I, DQ2556 + vasodilator (high dose); q l, vasodilator (high dose); *, significant difference from the control value (p < 0.05); #, significant difference from the DQ-2556 group (p < 0.05).
changes in the proximal tubular epithelium, from 30 min after the injection of DQ-2556. Desquamation of the brush borders and vesiculation of the cytoplasms were subsequently observed in the epithelial cells. Those findings have been included in the early changes of the kidney in rabbits, rats, and guinea pigs after administration of CER (Cojocel et al., 1983; Silverblatt et al., 1970; Silverblatt, 1982), and rabbits after cefotiam (Cojocel et al., 1988). Ischemia, too, has been reported to induce similar changes in the tubular epithelium of rats (Glaumann et al., 1975; Glaumann and Trump, 1975; Reimer et al., 1972; Venkatachalam et al., 1978). Thus, renal histological changes induced by cephalosporins are similar to ischemic changes. DQ-2556-induced changes are also similar to them and are suspected to be correlated to the constriction of the renal artery, being associated with early constriction of the intrarenal arteries. CER has been reported to inhibit the respiration of the mitochondria of the tubular epithelium (Tune and Hsu, 1990), and the inhibition has been shown to precede the earliest histological changes (Tune and Fravert, 1980a,b). Therefore, Tune and Fravert (1980b) consider that the inhibition of mitochondrial respiration is the most important molecular event in CER-induced renal injury. Respiratory inhibition of the mitochondria has also been demonstrated in ischemic renal injuries in rabbits (Tune and Hsu, 1990). Increased membrane permeability of the mitochondria, caused by peroxidation following to an increase in free radicals, has been reported to reduce production of ATP and to induce cell injury (Mason, 1986). On the basis of the above, it is possible that the key injury induced by DQ-2556, which
led to necrosis of the proximal tubular epithelium, is swelling of the mitochondria, which was seen in the early stages of renal injury. The Ca2+ channel blockers verapamil and nicardipine markedly, but not completely, inhibited tubular necrosis induced by DQ-2556. Verapamil and another Ca2+ channel blocker, nifedipine, have been reported to improve renal failure caused by contrast media, gentamicin or cyclosporine A (CSA) and to prevent renovasoconstriction induced by CSA (Rossi et al., 1989), norepinephrine, angiotensin, and vasopressin (Goldberg and Schrier, 1984). Respiratory inhibition of the mitochondria with the accumulation of Ca2’ has been reported in renal injury after norepinephrine-induced ischemia, and the inhibition is thought to be prevented by the above two vasodilators through retardation of the Ca*+ influx into the renal vascular cells and the tubular epithelium (Burke et al., 1984). From the above descriptions, it would be easy to attribute DQ-2556-induced-constriction of renal arteries and -necrosis of tubular epithelia to an increased influx of Ca*+ into the arterial smooth muscle cells and the tubular epithelial cells, and inhibition of renal injury to reduction of the influx by the Ca2+ channel blockers. Those mechanisms are thought to work in a large part for the DQ2556 nephrotoxicity in rats, but effects of other vasoconstrictive substances should be taken in consideration. In conclusion, DQ-2556-induced necrosis of the proximal tubular epithelial cells in the kidneys of rats is in a different manner from that of other cephalosporins generally showing close correlation between tubular necrosis and drug concentration in the renal cortex (Tune and Fravert, 1980a). DQ-
NEPHROTOXICITY
2556 simultaneously produced histological changes in the intrarenal arterial walls and reduced blood flow in the renal artery. Ca2+ channel blockers markedly inhibited both the tubular and vascular lesions, suggesting a correlation between tubular necrosis and vasoconstriction. ACKNOWLEDGMENTS The authors thank Dr. Shinjiro Hashimoto and Ms. Kumi Yoshida for measurement of drug concentrations in the kidneys.
REFERENCES Atkinson, R. M., Currie, J. P., Davis, B., Pratt, D. A. H., Sharpe, H. M.. and Tomich, E. G. (1966). Acute toxicity of cephaloridine, an antibiotic derived from cephalosporin C. Toxicol. Appi. Pharmacol. 8, 398-406. Bendirdjian, J.-P., Prime, D. J., Browning, M. C., Hsu, C.-Y., and Tune, B. M. (198 1). Additive nephrotoxicity of cephalosporins and aminoglycosides in the rabbit. J. Pharmacol. Exp. Thu. 218, 68 l-685. Bhan, R. D., Giacomelli, F., and Weiner, J. (1982). Adrenoceptor blockade in angiotensin-induced hypertension. Am. J. Pathol. 108, 60-7 1. Burke, T. J., Arnold, P. E., Gordon, J. A., Bulger, R. E., Dobyan, D. D., and Schrier, R. W. (1984). Protective effect of intrarenal calcium membrane blockers before or after renal ischemia. Functional, morphological. and mitochondrial studies. J. Clin. Invest. 74, 1830- 184 1. Cojocel, C., Smith, J. H., Maita, K., Sleight, S. D., and Hook, J. B. (1983). Renal protein degradation: A biochemical target of specific nephrotoxicants. Fundam. Appl. Toxicol. 3, 278-284. Cojocel, C., Gottsche, U., Tolle, K.-L., and Baumann, K. (1988). Nephrotoxic potential of first-, second-, and third-generation cephalosporins. Arch. Toxicol. 62, 458-464. Ejima. A., Hayano, T., Ebata, T., Nagahara, T.. Koda, H., Tagawa, H., and Furukawa. M. (1987). Synthesis and antimicrobial activity of cephalosporins with a 1-pyridinium substituent carrying a 5-membered heterccycle at the C-3 position. J. Antibiot. 40, 43-48. Fujimoto, T.. Otani. T., Nakajima, R., Une, T., and Osada, Y. (1986). In vitro activity of DQ-2556, a new cephalosporin. Antimicrob. Agents Chemother. 30, 6 1 l-6 13. Giacomelli. F., Anversa, P.. and Wiener, J. (1976). Effect of angiotensininduced hypertension on rat coronary arteries and myocardium. .4m. J. Pathol. 84, 11 I-138. Giese. J. (1964). Acute hypertensive vascular disease. 2. Studies on vascular reaction patterns and permeability changes by means of vital microscopy and colloidal tracer technique. Acta Pathol. Microbial. Stand. 62, 497515. Glaumann. B.. Glaumann. H., Berezesky, I. K., and Trump, B. F. (1975). Studies on the pathogenesis of ischemic cell injury. Il. Morphological changes of the pars convoluta (P, and P2) of the proximal tubule of the rat kidney made ischemic in vivo. Virchows Arch. B Cell Pathol. 19,28 l302. Glaumann, B., and Trump, B. F. (1975). Studies on the pathogenesis of ischemic cell injury. Ill. Morphological changes ofthe proximal pars recta tubules (P,) of the rat kidney made ischemic in vivo. Virchows Arch. B Cell Pathol. 19, 303-323. Goldberg, J. P., and Schrier, R. W. (1984). Effect of calcium membrane blockers on in vivo vasoconstrictor properties of norepinephrine, angiotensin II and vasopressin. Miner E/e&o&e Metab. 10, 178-I 83. Jones. G. M., and Hurley, J. V. (1985). The response of small blood vessels in rat skin and skeletal muscle to repeated application of histamine: A
OF DQ-2556 IN RATS
539
chemical. topographical, and electron microscopic study. J. Pathol. 146, 291-299. Joris. I., Majno, G.. and Ryan, G. B. (1972). Endothelial contraction in vivo: A study of the rat mesentery. Virchows Arch. B Cell Pathol. 12,7383. Joris. I.. and Majno, G. (198 1). Medial changes in arterial spasm induced by L-norepinephrine. Am. J. Pathol. 105, 2 12-222. Kai, M., Kanaide, H., Yamamoto, H., Kurozumi. T., Tanaka, K., and Nakamura, M. (198 1). The pathogenesis of small arterial lesions in nephrectomized rats by the administration of renin. Br. J. Exp. Pathol. 62, 136141. Kai-Yamamoto, M., Kanaide, H., Yamamoto, H., Kurozumi, T., Kato, K., and Nakamura, M. (1982). The production of small arterial lesions in nephrectomized rats by the administration of pure cathepsin D. Br. J. Exp. Pathol. 63, 501-505. Kerns, W. D., Arena, E., and Morgan, D. G. (1989). Role of dopaminergic and adrenergic receptors in the pathogenesis of arterial lesions induced by fenoldopam mesylate and dopamine in the rat. Am. J. Pathol. 135, 339-349. Kobori, K., Suzuki, K.. Yoshida, Y., and Ooneda, G. (1979). Light and electron microscopic studies on rat arterial lesions induced by experimental arterial contraction. Virchows Arch. A Pathol. Anat. Histol. 385, 29-39. Lawson, D. H., Macadam. R. F., Singh, H., Gavras, H., Hartz, S., Turnbull, D., and Linton, A. L. (1972). Effect of furosemide on antibiotic-induced renal damage in rats. J. Infect. Dis. 126, 593-600. Mason, J. (1986). The pathophysiology of ischemic acute renal failure. A new hypothesis about the initiation phase. Renal Physiol. 9, 129- 147. Reimer, K. A., Ganote, C. E., and Jennings. R. B. (1972). Alterations in renal cortex following ischemic injury. Ill. Ultrastructure of proximal tubules after ischemia or autolysis. Lab. Invest. 26, 347-363. Rossi, N. F., Churchill, P. C., Mcdonald, F. D., and Ellis, V. R. (1989). Mechanism of cyclosporine A-induced renal vasoconstriction in the rat. J. Pharmacol. Exp. Ther. 250, 896-901. Silverblatt, F., Turck, M., and Bulger, R. (1970). Nephrotoxicity due to cephaloridine: A light- and electron-microscopic study in rabbits. J. Infect. Dis. 122, 33-44. Silverblatt, F. (1982). Pathogenesis of nephrotoxicity of cephalosporins and aminoglycosides: A review of current concepts. Rev. Infect. Dis. 4(Suppl.), S360-S365. Tune, B. M., and Fravert, D. (1980a). Mechanisms of cephalosporin nephrotoxicity: A comparison of cephaloridine and cephaloglycin. Kidney Int. l&591-600.
Tune, B. M., and Fravert, D. (1980b). Cephalosporin nephrotoxicity. Transport. cytotoxicity and mitochondrial toxicity of cephaloglycin. J. Pharmacol. Exp. Ther. 215, 186-190. Tune, B. M.. and Hsu, C.-Y. (1990). The renal mitochondrial toxicity of cephalosporins: Specificity of the effect on anionic substrate uptake. J. Pharmacol. Exp. Ther. 252,65-67. Venkatachalam, M. A., Bernard, D. B., Donohoe, J. F.. and Levinsky, N. G. (1978). lschemic damage and repair in the rat proximal tubule: Differences among the S, , SZ, and S3 segments. Kidney Int. 14, 3 l-49. Vitullo, J. C.. Genity, R. G., and Khairallah. P. A. (1985). Cellular changes in mesenteric arteries and veins after acute perfusions of angiotensin II and vasoactive amines. Blood Vessel 22, 286-300. Wiener, J., and Giacomelli, F. (1973). The cellular pathology of experimental hypertension. VII. Structure and permeability ofthe mesenteric vasculature in angiotensin-induced hypertension. Am. J. Pathol. 72, 22 I-240. Wold, J. S.. and Tumipseed, S. A. (1980). The effect of renal cation transport inhibitors on the in vivo and in vitro accumulation and efflux of cephaloridine. f-i& Sci. 27, 2559-2564.
