w- .z “.3 . .9.» in. Y‘ -. .rnnz. Jr . . staffing?“ ant. Lw: x . I w .‘M\1 v. .luluv. .I‘ Zing-j. .‘-' .-, .1109." ." , . I'HESIL» , , C; {a ‘etfwfiflés' -' h; g. ,' n ‘1 ark/Lu.“ . in 'u A. Klv x} ‘1»- w ~ I‘- v- ar‘ 44\:‘.l~ $.- wa . I 7 ’1 “_ it»? &: HEAL; 3” dL '33"- «- \ fi This is to certify that the dissertation entitled BIOCHEMICAL MECHANISMS OF CEPHALORIDINE-INDUCED NEPHROTOXICITY presented by CHAD-H EN KUO has been accepted towards fulfillment of the requirements for 77,7) degree in 7>A43Wwéfl (”C/ 7;”‘(35/33‘? * cu 4/ / Vv//‘ij040fess;r Datejgéé’ 2244/7321 MSU is an Affirmative Action/Equal Opportunity Institution 0-12771 MSU LIBRARIES “ RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped below. BIOCHEMICAL MECHANISMS OF CEPHALORIDINE-INDUCED NEPHROTOXICITY By Chao-Hen Kuo A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Pharmacology and Toxicology 1982 ABSTRACT Biochemical Mechanisms of Cephaloridine-Induced Nephrotoxicity By Chao-Hen Kuo The purpose of this investigation was to determine the mechanism(s) responsible for cephaloridine-induced nephrotoxicity. Phenobarbital treatment induced mixed-function oxidases in rabbit renal cortex and also potentiated cephaloridine nephrotoxicity. 0n the other hand, treatment with piperonyl butoxide reduced cephaloridine nephrotoxicity in both rabbits and rats. A higher renal cortical concentration of cephaloridine was detected in phenobarbital-treated rabbits. Similarly, treatment with piperonyl butoxide decreased renal cortical accumulation of cephaloridine. Therefore, the potentiating and protective effects of phenobarbital and piperonyl butoxide, respectively, on cephaloridine nephrotoxicity might be due to increased and decreased renal cortical accumulation of the parent drug. Cephaloridine was found to be most nephrotoxic to rabbits, inter- mediate in toxicity to rats, and least toxic to mice. The relative susceptibility of these three species to GSH depletion in the renal cortex shortly after administration of cephaloridine paralleled species difference in nephrotoxicity of cephaloridine. In addition, pretreat- ment of animals with diethyl maleate potentiated cephaloridine Chao-Hen Kuo nephrotoxicity. Furthermore, cephaloridine markedly increased GSSG [coupled with the decreased GSH] in rat and rabbit renal cortex. These changes between GSH and GSSG could be the result of increased lipid peroxidation. Formation of conjugated diene in renal cortex was in- creased shortly following administration of cephaloridine. Removal of selenium and/or vitamin E from the diet potentiated cephaloridine nephrotoxicity suggesting that lipid peroxidation may be involved with cephaloridine nephrotoxicity. Alternatively, the high ratio of GSSG to GSH concentrations could be due to decreased NADPH concentrations in renal cortex. Low NADPH concentrations have been shown to inhibit several essential cellular functions. In addition, high concentration of GSSG alone will block many biochemical processes vital to the func- tions of renal cortical cells. Thus, cephaloridine-induced nephro- toxicity may be explained by one or a combination of the following three closely related potential changes, increased lipid peroxidation, in- creased intracellular GSSG and decreased NADPH. ACKNOWLEDGEMENTS I would like to express my appreciation to Dr. Jerry B. Hook for his enthusiastic assistance, helpful discussions and continuous finan- cial support. I would also like to thank the members of my guidance committee, Drs. Theodore M. Brody, w. Emmett Braselton, Jr., Robert A. Roth and Steven D. Aust for their suggestions and criticisms. All the members of Dr. Hook's laboratory have my acknowledgement for their discussions and suggestions. Finally, I want to thank Eli Lilly and Company for providing cephaloridine. ii TABLE OF CONTENTS Page ACKNOWLEDGEMENTS ------------------------------------------------- ii LIST OF TABLES --------------------------------------------------- vi LIST OF FIGURES -------------------------------------------------- ix INTRODUCTION ----------------------------------------------------- l Pharmacokinetics of Cephaloridine --------------------------- 3 Absorption and Distribution ---------------------------- 3 Excretion ---------------------------------------------- 4 Toxicity of Cephaloridine in Humans and Laboratory Animals-- 8 Renal Cortical Concentrations and Toxicity ------------------ ll Biochemical Mechanisms of Cephaloridine Nephrotoxicity ------ 15 Renal Drug-Metabolizing Enzymes ----------------------------- 20 Role of Glutathione in Cellular Protection Against Toxic Chemicals ---------------------------------------------- 22 Lipid Peroxidation and Its Role in Cellular Toxicity -------- 26 Objectives -------------------------------------------------- 3l METHODS ---------------------------------------------------------- 34 Animals ----------------------------------------------------- 34 Induction of Renal Microsomal Monoxygenases and Glutathione S-Transferases by Phenobarbital ------------------------ 34 Effects of Phenobarbital Pretreatment on Cephaloridine Toxicity ----------------------------------------------- 36 Effects of Piperonyl Butoxide Pretreatment on Cephaloridine Toxicity ----------------------------------------------- 36 Effects of Phenobarbital Pretreatment on Renal Cortical Up- take and Runout of Cephaloridine ----------------------- 37 Effects of Piperonyl Butoxide Pretreatment on Renal Cortical Accumulation of Cephaloridine -------------------------- 38 Effects of Phenobarbital Pretreatment on Renal Cortical Accumulation of Inulin and p-Aminohippurate Ig_Vivo and Cortical Slice Accumulation of p-Aminohippurate and Tetraethylammonium Ig_Vitro ---------------------------- 38 Effects of Piperonyl Butoxide Pretreatment on Renal Cortical Accumulation of p-Aminohippurate, Tetraethylammonium and Inulin --------------------------------------------- 4O Depletion of Renal Glutathione Concentration and Nephrotoxi- city of Cephaloridine in Rabbits, Rats and Mice -------- 4O TABLE OF CONTENTS (continued) Page METHODS (con'd) Dose-Dependent Effect of Cephaloridine, Cephalothin and Gentamicin on Tissue GSH Concentration ----------------- 41 Effect of Cephaloridine on Tissue Reduced and Oxidized Glutathione (GSH and GSSG) and Conjugated Dienes ------- 42 Effect of Vitamin E and/or Selenium Deficiency on Cephalori- dine Toxicity in Rats ---------------------------------- 42 Analytical Methods ------------------------------------------ 45 Determination of PAH and TEA Accumulation and Gluconeo- genesis in Renal Cortical Slices ------------------ 45 Determination of Blood Urea Nitrogen and Serum Glutamic Pyruvic Transaminase Activity --------------------- 46 Determination of Tissue Water Content and Reduced Glu- tathione (GSH) and Oxidized Glutathione (GSSG) Concentration ------------------------------------- 47 Measurement of Conjugated Diene in Microsomes ---------- 48 Determination of Serum and Tissue Cephaloridine Concen- tration ------------------------------------------- 48 Determination of Cephaloridine Concentrations in Renal Cortical Slices and Incubation Media -------------- 50 Determination of Potential GSH Conjugates of Cephalori- dine or Its Metabolite(s) ------------------------- 50 Histology --------------------------------------------------- 52 Statistics -------------------------------------------------- 53 RESULTS ---------------------------------------------------------- 54 Induction of Renal Microsomal Monooxygenases and Glutathione S-Transferase Activity --------------------------------- 54 Inhibition of Renal Glutathione S-Transferase Activities by Cephaloridine In Vitro --------------------------------- 54 Effect of Phenobarbital on Cephaloridine Nephrotoxicity ----- 57 Effect of Piperonyl Butoxide Pretreatment on Cephaloridine Nephrotoxicity ----------------------------------------- 57 Effect of Sex Difference on Cephaloridine Toxicity ---------- 63 Effect of Phenobarbital on Renal Cortical Accumulation of Cephaloridine In Vivo ---------------------------------- 63 Effect of Phenobarbital Pretreatment on Cephaloridine Uptake and Runout in Rabbit Renal Cortical Slices ------------- 67 Effect of Piperonyl Butoxide Pretreatment on Renal Cortical Accumulation of Cephaloridine In Vivo and In Vitro ----- 67 Effect of Phenobarbital Pretreatment—on Rabbit REnal Corti- cal Accumulation of PAH and Inulin In Vivo and PAH and TEA In Vitro ------------------------------------------- 74 iv TABLE OF CONTENTS (continued) Page RESULTS (con'd) Effect of Piperonyl Butoxide on Rabbit Renal Cortical Accu- mulation of PAH and Inulin Ig_Vivo and PAH and TEA In Vitro ----------------------------------------------- 74 Kidney and Liver Toxicity in Rabbits, Rats and Mice Forty- Eight Hours After A Single Dose of Cephaloridine ------- 81 Effect of Cephaloridine on Kidney and Liver Glutathione (GSH) and Water Content in Rabbits, Rats and Mice ------ 85 Effect of Diethyl Maleate Pretreatment on Cephaloridine Nephrotoxicity ----------------------------------------- 91 Effect of Cephaloridine, Cephalothin and Gentamicin on Tissue GSH in Sprague-Dawley Rats ---------------------- lOO Formation of Glutathione Conjugate(s) of Cephaloridine or Its Metabolite(s) Ifl_Vitro ----------------------------- lOO Effect of Cephaloridine on Renal Cortical and Hepatic GSH and GSSG in Rabbits and Rats --------------------------- lOO Effect of Cephaloridine on Renal Cortical and Hepatic Con- jugated Dienes ----------------------------------------- 108 Effect of Vitamin E and/or Selenium Deficiency on Cephalori- dine Toxicity in Rats ---------------------------------- l08 DISCUSSION ------------------------------------------------------- 127 CONCLUSION ------------------------------------------------------- 154 BIBLIOGRAPHY ----------------------------------------------------- 156 Table ll 12 l3 14 LIST OF TABLES Composition of Basal Diet .............................. Mineral Mix, Hubbell-Mendel-Nakeman -------------------- Effect of Phenobarbital Treatment on Renal Microsomal Monooxygenases in Rats and Rabbits ..................... Effect of Phenobarbital Treatment on Renal and Hepatic Glutathione S-Transferase Activity in Rats and Rabbits- In Vitro Effect of Cephaloridine on Rat Renal Cortical ETUtatfiione S-Transferase Activity ..................... Effect of Phenobarbital Pretreatment on Cephaloridine Toxicity in Rabbits .................................... Effect of Piperonyl Butoxide Pretreatment on Cephalo- ridine Nephrotoxicity in Sprague-Dawley Rats ----------- Effect of Piperonyl Butoxide Pretreatment on Cephalo- ridine Nephrotoxicity in Rabbits ....................... Cephaloridine Toxicity in Male and Female Rats --------- Effect of Phenobarbital Pretreatment on Renal Cortical Accumulation of Cephaloridine in Rabbits ............... Effect of Phenobarbital Pretreatment on Renal Cortical Accumulation of Cephaloridine in Rats .................. Effect of Phenobarbital Pretreatment on Cephaloridine Accumulation in Rabbit Renal Cortical Slices ----------- Effect of Piperonyl Butoxide Pretreatment on Cephalori- dine Accumulation by Rabbit Renal Cortex --------------- Effect of Piperonyl Butoxide Pretreatment on Cephalori- dine Uptake in Rabbit Renal Cortical Slices ------------ vi Page 43 44 55 56 58 61 62 64 65 66 68 69 72 73 LIST OF TABLES (continued) Table 15 l6 T7 18 I9 20 21 22 23 24 25 26 27 28 29 Page Effect of Phenobarbital Pretreatment on Renal Cortical Accumulation of PAH and Inulin in Rabbits -------------- 75 Effect of Phenobarbital Pretreatment on Accumulation of p-Aminohippurate and Tetraethylammonium in Rabbit Renal Cortical Slices ---------------------------------------- 76 Effect of Piperonyl Butoxide Pretreatment (30 min) on PAH and Inulin Accumulation in Rabbit Renal Cortex ----- 77 Effect of Piperonyl Butoxide Pretreatment (90 min) on PAH and Inulin Accumulation in Rabbit Renal Cortex ----- 78 Effect of Piperonyl Butoxide Pretreatment (90 min) on p-Aminohippurate and Tetraethylammonium Accumulation in Rabbit Renal Cortical Slices --------------------------- 79 Effect of Piperonyl Butoxide Pretreatment (45 min) on p-Aminohippurate and Tetraethylammonium Accumulation in Rabbit Renal Cortical Slices --------------------------- 8O Cephaloridine Toxicity in Rabbits ---------------------- 82 Cephaloridine Toxicity in Rats ------------------------- 83 Cephaloridine Toxicity in Mice ------------------------- 84 Effect of Cephaloridine Treatment on Mouse Renal Corti- cal GSH Concentration ---------------------------------- 89 Water Content in Rabbit and Rat Renal Cortex and Liver Following Cephaloridine Administration ----------------- 90 Effect of Diethyl Maleate Treatment on Rat Renal Corti- cal GSH Content ---------------------------------------- 96 Effect of Diethyl Maleate Pretreatment on Cephaloridine Nephrotoxicity in Rabbits ------------------------------ 97 Effect of Diethyl Maleate Pretreatment on Cephaloridine Nephrotoxicity in Rats --------------------------------- 98 Effect of Diethyl Maleate on Cephaloridine Nephrotoxi- city in Sprague-Dawley Rats ---------------------------- 99 vii LIST OF TABLES (continued) Table 30 31 32 33 Page Effect of Gentamicin Treatment on Sprague-Dawley Rat Tissue GSH Concentrations ------------------------------ lO3 Effect of Cephaloridine Treatment on Reduced and Oxi- dized Glutathione Concentrations in Rat Kidneys and Livers ------------------------------------------------- 106 Effect of Cephaloridine Treatment on Reduced and Oxi- dized Glutathione Concentrations in the Rabbit Kidneys and Livers --------------------------------------------- lO7 Formation of Conjugated Dienes in Rat Renal Cortex and Liver Following Cephaloridine Administration ----------- lO9 viii Figure 10 11 12 13 14 15 LIST OF FIGURES Page The structure of cephaloridine ------------------------- 2 The structure of paraquat ------------------------------ 30 The cyclic reduction-oxidation of paraquat and concomi- tant formation of superoxide --------------------------- 32 The proposed cyclic reduction and oxidation of cepha- loridine ----------------------------------------------- 33 Effect of phenobarbital pretreatment on blood urea ni- trogen and renal cortical slice accumulation of PAH and TEA in rats treated with cephaloridine ----------------- 59 Effect of phenobarbital pretreatment on cephaloridine runout in rabbit renal cortical slices ----------------- 70 Time course of rabbit tissue GSH following a single sc administration of cephaloridine ------------------------ 86 Time course of rat tissue GSH following a single ip administration of cephaloridine ------------------------ 88 Dose-dependent depletion of rabbit tissue GSH following cephaloridine administration --------------------------- 92 Dose-dependent depletion of rat tissue GSH following cephaloridine administration --------------------------- 94 Dose-dependent depletion of Sprague-Dawley rat tissue GSH following cephaloridine administration ------------- 101 Dose-dependent depletion of Sprague-Dawley rat tissue GSH following cephalothin administration --------------- 104 Effect of cephaloridine on rabbit renal cortical con- jugated dienes ----------------------------------------- 110 Effects of vitamin E and/or selenium deficiency on cephaloridine toxicity in rats ------------------------- 112 Effects of vitamin E and/or selenium deficiency on blood urea nitrogen in rats after cephaloridine -------- 114 ix LIST OF FIGURES (continued) Figure 16 17 18 19 20 21 22 23 24 Page Effects of vitamin E and/or selenium deficiency on glu- coneogenesis by renal cortical slices in rats after cephaloridine ------------------------------------------ 116 Effects of vitamin E and/or selenium deficiency alone on rat renal histologic structure ---------------------- 118 Effects of vitamin E and/or selenium deficiency on rat renal histologic structure after 500 mg/kg of cepha- loridine ----------------------------------------------- 120 Effects of vitamin E and/or selenium deficiency on rat renal histologic structure after 1000 mg/kg of cepha- loridine ----------------------------------------------- 123 Effects of both vitamin E and selenium deficiency on rat renal histological structure after 1000 mg/kg of cephaloridine ------------------------------------------ 125 Schematic diagram of the proposed transport of cepha- loridine in renal proximal tubular cells --------------- 136 Possible pathways for the formation of glutathione con- jugate(s) of cephaloridine or its metabolite(s) -------- 140 Proposed mechanism for cephaloridine nephrotoxicity in- volving lipid peroxidation ............................. 146 Schematic diagram of the proposed mechanisms for cepha- loridine-induced nephrotoxicity ------------------------ 155 INTRODUCTION In 1945 Brotzu isolated a fungus, Cephalosporium acremonium, from the sea near a sewage outlet in Sardinia. Crude extract from cultures of this fungus was found to cure typhoid fever and Brucella infection. A culture of this microorganism was sent to Oxford, where seven anti- biotic substances were isolated (Florey, 1955). Three of these were cephalosporin N, a new type of penicillin which is active against gram- negative and gram-positive microorganisms; cephalosporin P, which has a steroid structure and is active only against gram-positive microorga- nisms; and cephalosporin C which is less potent than cephalosporin N but possesses the same range of antimicrobial effectiveness. The nucleus of cephalosporin C is 7-aminocephalosporanic acid and with the addition of side chains, it became possible to produce semisynthetic cephalosporins with antibacterial activity much greater than that of the parent sub- stance. Cephaloridine (Figure 1), 7-[(2-thieny1)acetamidOJ-B-(l-pyridyl- methyl)-3-cephem-4-carboxy1ic acid betaine, is one of the semisynthetic derivatives of cephalosporin C (Loder et_al,, 1961). This antibiotic has a broad spectrum of activity against many gram-positive and gram- negative microorganisms (Murdoch gt_al,, 1964; Turck gt 21-, 1967). It has no action against fungi, protozoa, or helminths and has only low .mcwuwgopmgnmo mo weaposcpm mgp .P mgamwm o0/o\o 2 Qmmzéfizlmlazé 3 activity against Mycobacterium tuberculosis. The antibiotic is highly bactericidal, particularly against gram-negative organisms. Pharmacokinetics of Cephaloridine Absorption and Distribution. Cephaloridine is poorly and irre- gularly absorbed from the gastrointestinal tract (Kislak et_al,, 1966). Serum concentration of the drug in humans after an oral dose of l g in the fasting state never achieved a level of 0.5 pg/ml at any time within 2 hours. In contrast, high serum concentrations (10-25 pg/ml) were achieved within 15 minutes when the drug was given parenterally. Welles 1. (1966) reported that the serum half-life of cephaloridine was 0.5 _e_t hr in the dog after a single intravenous injection. A similar serum half-life was also observed when the drug was administered intramuscu— larly, indicating a rapid absorption from the injection site. A rela- tively longer serum half-life was observed in humans. Kirby et_al, (1971) reported a serum half-life of 1.12 hr in healthy volunteers. This value is very close to the 1.5 hr serum half-life in patients with normal kidney function (Kabins and Cohen, 1966). The serum half-life of cephaloridine is prolonged to 20-23 hr in patients with renal impairment (Kunin and Atuk, 1966; Kirby gt_al,, 1971). Unlike cephalothin, another cephalosporin, cephaloridine is less protein-bound, about 20% compared to 65% for cephalothin (Kirby et 31,, 1971). Cephaloridine is distri- buted in a variety of tissues and also is found in the fetus and milk (Welles gt_al,, 1966). Kidney tissue has the highest concentration of the drug; other tissues contain only one-tenth or less of the 4 concentration in the kidney (Welles et 11., 1966). Most of the drug in the kidney is found in the cortical region (Tune §t_al,, 1974). Excretion. Only a small percentage of cephaloridine is excreted through the bile (Mandell, 1973). Biliary concentrations of the drug were found to be dose-unrelated and also to be lower than serum concen- trations (Nishida gt_al,, 1976). Elimination of cephaloridine occurs mainly through the kidney. In dogs, approximately 92% of intravenously administered drug was recovered in the urine within 6 hours (Welles gt 31,, 1966). More than 80% of the drug was excreted in the urine within 24 hours in humans (Kirby gt_al,, 1971). Child and Dodds (1966) first investigated the mechanism of urinary excretion of cephaloridine in various species of animals. Because of low protein binding, most cephaloridine appeared to be excreted by glomerular filtration (Child and Dodds, 1966). In addition, using the method of Sperber (1948), Child and Dodds found that excretion of cephaloridine by chickens was higher on the infused side and these authors suggested that tubular secretion of cephaloridine occurred in hen kidneys. However, this may not be true for other species. Renal clearances of cephaloridine in anesthetized rabbits, cats, dogs and monkeys were 1 to 5 ml/kg/min, very closed to the value found in humans, 3 ml/kg/min (Kirby §t_al,, 1971). The ratios of cephaloridine to creatinine clearance in these animals ranged from 0.30 to 1.20, indicating that there was not secretion but a small reabsorption of cephaloridine. A saturable tubular reabsorption in humans was also suggested (Arvidson gt_al,, 1979). Although renal clearance studies did not provide evidence of renal tubular secretion of cephaloridine, it is possible that bidirectional tubular transport of 5 the drug may occur. In order to investigate this possiblity, Child and Dodds (1967) used stop-flow techniques (Malvin gt_al,, 1958) to re- examine nephron transport of cephaloridine in dogs and rabbits. The results indicated that cephaloridine was neither secreted nor reabsorbed by the renal tubules. Furthermore, probenecid, an inhibitor of organic acid transport, was shown to block the active secretion of cephaloridine in the hen kidneys (Child and Dodds, 1966), but this inhibitor had no effect on the stop flow pattern of cephaloridine in dogs and rabbits (Child and Dodds, 1967) nor the renal clearance of cephaloridine in dogs (Welles §t_al,, 1966). These results led Child and Dodds to conclude that cephaloridine might not be handled by an organic acid transport system, at least in dogs and rabbits. However, several lines of evi- dence argued against this suggestion. Child and Dodds (1967) found that prior administration of probenecid and other inhibitors of organic acid transport prevented the renal lesion caused by cephaloridine (cepha- lorodine-induced nephrotoxicity will be described in detail later) in mice. This protecting effect was also demonstrated in rabbits (Tune, 1972). Furthermore, cephalosporin drugs are structurally related to the penicillins (Loder e__al,, 1961), which have been shown to be actively secreted by renal organic acid transport (Beyer, 1950). In addition, cephalothin was found to be transported by an organic acid transport system (Lee gt 31,, 1963). All this information strongly suggested a possible connection between cephaloridine and organic acid transport. Interestingly, all the investigations on the renal handling of cepha- loridine were directed solely towards cephaloridine secretion and 6 completely neglected the possibility that cephaloridine might be con- centrated in the renal proximal tubular cells in the absence of secre- tion. This possibility was confirmed a few years later by Tune and his coworkers (Tune, 1972; Tune and Fernholt, 1973). Tune (1972) studied renal uptake of cephaloridine in rabbits by giving single subcutaneous injections of the drug and measuring its concentration in serum, renal cortex and medulla 30 minutes later. Cortex/serum ratios of cepha- loridine were approximately 8 times higher than those of inulin, sug- gesting that renal cortical cephaloridine concentration was signifi- cantly greater than that which could be accounted for by glomerular filtration alone. Prior administration of probenecid or benzylpeni- cillin markedly reduced cortex/serum ratios of cephaloridine. Further- more, Tune and Fernholt (1973) were able to show that the uptake of cephaloridine by renal cortical slices was oxygen dependent and inhi- bited by dinitrophenol, indicating that the transport of cephaloridine into renal cortical tissue was an active process. Thus, although there is no significant cephaloridine secretion into the urine, there is active cortical cephaloridine uptake by organic acid transport. In order to understand why cephaloridine is accumulated in renal cortex, Tune and his coworkers (1974) conducted a series of elegant experiments. These investigators ligated the ureter to interrupt the movement of PAH from proximal tubular fluid into the urine subsequently retarding PAH movement from cell water into luminal fluid. Six minutes of ureteral ligation resulted in a doubling of renal cortical PAH concentration but had no effect on renal cortical cephaloridine concentration, indicating 7 cephaloridine movement from cell water to luminal fluid was much slower than PAH movement. Furthermore, abrupt inhibition of transport by an intravenous bolus of probenecid resulted in a rapid loss of PAH from proximal tubular cells. However, intravenous administration of pro- benecid only resulted in a slow decrease of the renal cortical concen- tration of cephaloridine. These results and those from other investi- gators led Tune (1975) to conclude that cephaloridine, like PAH, was actively transported into the proximal tubular cells at the peritubular cell membrane, but unlike PAH, cephaloridine did not move readily across the luminal cell membrane into the tubular fluid. It can be concluded from these studies that although there is no net secretion of cepha- loridine, cephaloridine does depend on active transport for uptake into cortical tubular cells and this transport can be blocked by organic acid transport inhibitors. The role of organic acid transport in renal cortical accumulation of cephaloridine was further confirmed by Wold and Turnipseed (1978) who investigated the accumulation of cephaloridine in the renal cortex of rabbits of various ages. The cortical concentration of cephaloridine rose from newborn to adult levels by approximately one month of age. This pattern correlated well with the development of organic acid transport studied with the classical substrate PAH (Hirsch and Hook, 1970). Furthermore, pretreatment of immature rabbits with procaine penicillin G has been shown to stimulate the development of organic acid transport (Hirsch and Hook, 1970). Similarly, penicillin G treatment stimulated the ability of immature rabbit renal cortical cells to accumulate cephaloridine (Wold and Turnipseed, 1978). 8 Even though it is known that the egress of cephaloridine from renal cortical cells is much slower than PAH, the process for cephaloridine efflux was not studied until recently. Hold and Turnipseed (1980) found that pretreatment of rabbits with cyanine, an inhibitor of organic base transport, led to retention of cephaloridine in the renal cortex. In addition, the efflux of cephaloridine from preloaded renal cortical slices was significantly slowed by the presence of cyanine in the efflux media. These results indicated that cephaloridine depended on, at least in part, a base transport step for exit from proximal tubular cells. Toxicity of Cephaloridine in Humans and Laboratory Animals Cephaloridine and other cephalosporins share several toxicities, which are mainly allergic reactions. However, the main adverse effect of cephaloridine is nephrotoxicity. Many studies have documented serious and even fatal nephrotoxicity from cephaloridine in humans with previously normal kidneys who received doses greater than 6 grams daily (Kaplan gt 21,, 1968; Hinman and Wolinsky, 1967). Although treatment with doses of 4 g or less daily in patients with normal renal function does not result in detectable renal impairment (Fair, 1972), the risk of nephrotoxicity by cephaloridine may be enhanced by other agents such as aminoglycosides and diuretics (Lawson gt_al,, 1970; Kleinknecht gt_al,, 1974). In addition, patients who have illnesses that directly or indirectly affect renal function may develop nephrotoxicity with doses of 4 g cephaloridine daily or less (Mandell, 1973). Nephrotoxicity of cephaloridine has been demonstrated in many species of laboratory animals (Atkinson gt_al,, 1966), especially monkeys and rabbits (Perkins gt_ 1., 1968). Cephaloridine-treated 9 rabbits showed dose-dependent alteration in renal function. Those given 50-100 mg/kg per day had no significant change in creatinine clearance and no glycosuria or proteinuria, whereas a dosage of 200 mg/kg per day caused marked changes in renal function. All animals receiving this dose developed proteinuria and some animals had concomitant glycosuria. Rabbits given 500 mg/kg of cephaloridine per day, developed glycosuria and severe proteinuria (Perkins g__al,, 1968). Alterations in renal function were compatible with histopathological findings. Silverblatt gt_gl, (1970) reported that rabbits receiving 200 mg/kg developed alteration in the proximal tubular brush border at one hour and frank necrosis was evident at 16 hours. Early changes seen with the elec- tron microscope were 1055 of microvilli and the disappearance of struc- tures associated with endocytosis. Later, disorganization of lateral interdigitations of cell membranes and mitochondrial swelling were observed. Necrosis did not result from 50 mg/kg cephaloridine treat- ment. Proximal tubular necrosis developed in some rabbits receiving 100 mg/kg of cephaloridine. In addition, uptake of horseradish peroxidase was blocked in damaged tubules one hour after administration of cepha- loridine, indicating interference with endocytosis. These data demon- strated that in rabbits, cephaloridine produced a dose-dependent lesion of the proximal tubules that resulted in early disruption of structure and function of the cell membrane. Perkins et_gl, (1968) also reported a dose-related nephrotoxicity in Rhesus monkeys. Dosage of 100 mg/kg per day resulted in little change in renal function. Monkeys given 200 mg/kg per day had a marked drop in PAH clearance and a concomitant marked rise in BUN and serum creatinine. Proximal tubular alteration 10 and necrosis were observed in these animals (Perkins gt_al,, 1968). A similar finding was also reported by Atkinson et_al, (1966). In addi- tion to rabbits and monkeys, renal tubular necrosis was seen in rats and mice treated with higher dosage of cephaloridine (Welles §t_al,, 1966; Atkinson et_al,, 1966). It has been shown that when the kidney is affected by the action of certain toxic compounds, enzymes are usually released from the damaged cells and pass into the blood, urine and other extracellular fluids. Measurements of such enzyme activities and their characterization may provide a useful index of kidney damage. For instance, lactate dehy- drogenase, alkaline phosphatase, muramidase, acid phosphatase and B- glucosidase have been shown to be elevated in the urine following damage to the kidney in animals and man. Ngaha and Plummer (1977) monitored the change in the rat urinary lactate dehydrogenase, alkaline phospha- tase and muramidase activities following cephaloridine administration; urinary lactate dehydrogenase and muramidase were elevated within 12 hours following subcutaneous administration of 2 g/kg cephaloridine and reached a peak at 60 hours. A concomitant decrease in kidney lactate dehydrogenase and alkaline phosphatase was also determined. Serum enzymes were also studied to determine the possible sources of urinary enzymes in cephaloridine-induced nephron damage. The results indicated that serum enzyme levels were not much affected. Raab and Moerth (1976) also reported an increased excretion of alkaline phosphatase, leucine aminopeptidase and lactate dehydrogenase after the administration of cephaloridine to rats at a dose of 250 mg/kg. In addition, Sack (1976) observed increased urinary enzyme excretion as well as an increased rate ll of excretion of renal epithelial cells after the administration of cephaloridine at a dose of 500 mg/kg per day to Wistar rats. In con- trast, cephaloridine appeared to have different effects on lysosomes. Following 2 g/kg, lysosomal enzyme acid phosphatase in the urine was suppressed (Wright gt_§l,, 1974). This decrease was shown to be due to the stabilization of lysosomal membranes by cephaloridine (Fry gt 31., 1975; Ngaha gt_al,, 1979). Later, cephaloridine was shown to inhibit a lysosomal membrane-bound phospholipase 2, the enzyme which digests phospholipids of lysosomal membrane. Such an inhibition may explain the cephaloridine-induced stabilization of rat kidney lysosomes (Fry and Plummer, 1979). More recently, the toxicity of cephaloridine in rabbits has been characterized by other techniques, and the antibiotic has been shown to produce changes in transport and metabolism of the kidney as measured by the ability of renal cortical slices to accumulate the anion PAH or the cation TEA and to perform gluconeogenesis (Wold et_al., 1979). Renal Cortical Concentrations and Toxicity Several lines of evidence have suggested a close correlation be- tween cephaloridine nephrotoxicity and renal cortical concentration of cephaloridine. The early report of the nephrotoxicity of cephaloridine by Welles et_al, (1966) showed an unusually high concentration in the renal cortex, although they did not suggest that this unique high con- centration caused renal injury. As described previously, cephaloridine nephrotoxicity in rabbits and mice was prevented by probenecid (Child 12 and Dodds, 1966; Tune, 1972). This protection appeared to be associated with decreased cephaloridine concentration in renal cortex (Tune gt al., 1977a). Tune et_al, (1977a) examined the effect of various doses of probenecid on the nephrotoxicity and renal cortical concentration of cephaloridine in rabbits. Probenecid at doses of 20 to 60 mg/kg pro- duced a dose-related decrease in the severity of cephaloridine-induced renal cortical necrosis. Interestingly, probenecid also caused a dose- related decrease in renal cortical concentration of cephaloridine. A similar effect of probenecid on cephaloridine concentration and nephro- toxicity was demonstrated in guinea pigs (Tune §t_al,, 1977a). Additional evidence for this causal relationship between renal cortical concentration and nephrotoxicity of cephaloridine came from the studies by Fleming and Jaffe (1967) and Wold gt_al. (1977a; 1978). Fleming and Jaffe (1967) first reported that newborn rabbits were less susceptible to cephaloridine nephrotoxicity. This observation was extended by Wold gt al, (1977a), who investigated the susceptibility of rabbits of various ages to cephaloridine nephrotoxicity. Cephaloridine administered at doses of 200 or 400 mg/kg produced severe renal cortical tubular necrosis and elevations of BUN and creatinine in rabbits 30 days of age and in adult rabbits. However, doses as high as 800 mg/kg pro- duced no evidence of cortical tubular necrosis or alteration in renal function in rabbits 5 days of age. The relative resistance of the newborn rabbit kidney to cephaloridine nephrotoxicity did not appear to be a generalized property of the immature kidney since mercuric chlor- ide, a nephrotoxicant which also produced cortical tubular necrosis, has 13 been shown to be equally toxic in the immature.and adult rabbits (Wold gt_al,, 1977a). The development of susceptibility to cephaloridine nephrotoxicity paralleled development of the ability of renal cortical cells to accumulate cephaloridine (Wold gt al,, 1978). Further evidence for the link between renal cortical concentration and nephrotoxicity of cephaloridine was obtained from experiments in which rabbits were pre- treated with PAH or penicillin G. Pretreatment with either PAH or penicillin markedly enhanced the susceptibility of animals of 15 to 18 days of age to cephaloridine nephrotoxicity (Wold gt_al,, 1977a) and similarly enhanced the uptake of cephaloridine in renal cortex (Wold gt al,, 1978). Cephaloridine nephrotoxicity varied among the species. Atkinson at 31. (1966) calculated the N050, which they defined as the nephrotoxic dose for each species producing histologically evaluable changes at 48 hr in 50% of the animals treated. ND50 was lowest in the rabbit (90- 140 mg/kg) and increased in other species as follows: monkey, 300 mg/kg; guinea pig, 400-700 mg/kg; rat, 1000-1400 mg/kg; and mouse, 600- 3100 mg/kg. The dog and the cat were not affected by 1000 mg/kg. 0n the basis of this information, Tune (1975) measured renal cortical concentrations of cephaloridine in rabbits, guinea pigs and rats follow- ing s.c. administration of 100 mg/kg of cephaloridine. Cortical con- centration was highest in the rabbit and lowest in the rat, which was consistent with the severity of cephaloridine nephrotoxicity in these species studied. The nephrotoxicity of cephaloridine exhibited a threshold in rabbits and guinea pigs (Tune et_al,, 1977a). In both species, cortical 14 concentrations of cephaloridine above 2000 pg/g at 3 hours following cephaloridine administration produced significant tubular necrosis, while concentrations remaining below 1000 pg/g caused little or no cell damage. This finding was extended by the studies of Hold (1981). He found that renal cortical function (e.g. organic acid and base transport and gluconeogenesis) began to decline when cortical concentrations of cephaloridine were approximately 1200 pg/g. When cortical concentra- tions were above 1500 pg/g, marked nephrotoxicity evidenced by BUN elevation was observed. Although all cephalosporins possess the carboxyl functional group of the beta-lactam ring necessary for interaction with renal organic anion transport, cephaloridine is unique in that it possesses a cationic functional group in the quaternary nitrogen of the pyridinium substi- tution on the beta-lactam ring. Because of this cationic property of cephaloridine, Wold et_al, (1979; 1980) suggested that cation transport, like anion transport, might also play a role in determination of renal cortical concentrations of cephaloridine. These investigators found that pretreatment of rabbits with the cation transport inhibitors cyanine 863 or mepiperphenidol (Peters gt_al,, 1955; Rennick gt _l,, 1956; Kandel and Peters, 1957), markedly enhanced the nephrotoxicity of cephaloridine, but had no effect on the nephrotoxicity of cefazolin, a cephalosporin which has no cation functional group. Furthermore, administration of cyanine at intervals of up to 3 hr after administra- tion of cephaloridine was also shown to enhance cephaloridine nephro- toxicity, suggesting that the inhibition of cation transport might 15 result in a decreased efflux of cephaloridine out of cells of the renal cortex. Examination of the effect of cyanine pretreatment on cortical concentrations of cephaloridine as well as evaluation of the effect of cyanine on the efflux of cephaloridine jn_vitrg_clearly indicated that although peak concentrations of cephaloridine obtained jn_vjv9_were unaltered the renal cortical concentrations were prolonged and efflux jg_ vitrg_was slowed by inhibition of cation transport (Wold and Turnipseed, 1980). Biochemical Mechanisms of Cephaloridine Nephrotoxicity Since cephaloridine was recognized as being capable of producing renal cortical damage, several biochemical mechanisms have been proposed to explain this toxic effect. Boyd et_al, (1971; 1973) first exten- sively investigated the mechanism of cephaloridine nephrotoxicity. Cephaloridine solution was shown to contain polymeric macromolecules. Using Sephadex column chromatography and ultrafiltration, Boyd et_al, (1971) obtained 4 fractions from commercial cephaloridine preparations. The first was found in only minute amounts and had a molecular weight (MW) greater than 5000; this fraction was inconstant and had the charac- teristics of a polypeptide-polymer complex. The next 3 fractions were constantly present from batch to batch. The first of these, polymer I (5000 1 MW‘: 3500) occurred as 45% of starting material. The second, polymer II (3500 :_MW.: 1000) formed 26% and the final fraction (1000 3 MW) was present as 29% of starting material. Because of inconstant presence and minute amounts, these investigators did not study the first fraction. Boyd and coworkers (1973) studied the remaining fractions and attempted to determine which fraction(s) was responsible for cepha- loridine nephrotoxicity. Different degrees of nephrotoxicity were 16 observed with these fractions. Renal tubular damage was found to be least with the high molecular weight polymer (polymer I), greater with the medium-sized polymer (polymer II) and greatest with the fraction with molecular weight less than 1000 (polymer III). The polymer III fraction contained monomers and dimers of cephaloridine and probably degradation products. Since pyridine may be released from cephaloridine and retained in the polymer III fraction, these investigators also studied the nephrotoxicity of pyridine. Pyridine failed to cause any significant renal tubular cell damage at equimolar doses. In addition, reconstitution by mixing these three fractions (polymer I, II and III) only led to about half of the renal tubular damage caused by the starting material, indicating that the untested polypeptide-polymer complex fraction coud play some part in causing nephrotoxicity. How- ever, these investigators (Boyd gt 11., 1973) did not favor this possi- bility, because not all batches of commercial cephaloridine used in their studies contained this fraction, whereas all batches appeared to be equally nephrotoxic. Although none of the fractions was solely responsible for cephaloridine nephrotoxicity, the severity of nephro- toxicity was decreased with an increase in the degree of polymerization. The greatest tubular damage resided with the fraction (polymer III) which had a strong tendency to polymerize; less damage occurred with polymer II which was probably halfway through the polymerizing process and least damage occurred with polymer I which was the most stable fraction, being nearest to the end of the polymerizing process. These possibilities led Boyd gt El: (1973) to suggest that the process of polymerization might be responsible for the tubular damage; this process 17 was more likely to occur in concentrated solutions such as those in the lumen of the proximal tubule. Such an effect might also occur in the cytoplasm of tubular cells. These investigators did not exclude the possibility of precipitation of cephaloridine from concentrated or supersaturated solutions. Precipitated antibiotic might complex with protein to form haematoxyphilic granular deposits. The second possible biochemical mechanism of cephaloridine nephro- toxicity was proposed by Tune at al. (1979), who suggested that cepha- loridine first interfered with renal cortical mitochondrial function which then led to cell damage and death. In order to test their hypo- thesis, these investigators examined the effect of cephaloridine on respiration by mitochondria in intact cells (renal cortical tubules) and isolated renal cortical mitochondria (Tune gt 21., 1979; Tune and Fravert, 1980). In the isolated mitochondria studies, the effect of cephaloridine on respiration was examined under two conditions: (a) j__ vitr9_exposure: respiration studied before and after exposure of isolated mitochondria to cephaloridine; (b) jg_vjvg_exposure: respira- tion studied in mitochondria isolated from animals that received cepha- loridine prior to sacrifice. In these studies, respiration by renal cortical mitochondria was studied with each of three substrate groups including glutatmate plus maleate, succinate, and tetraethylphenylene- diamine plus ascorbate. The respiratory control ratio (RCR), which is the ratio of the higher rate of oxygen consumption in the presence of ADP (state 3) to the lower rate after the consumption of ADP (state 4), was used as a functional measure of the integrity of mitochondria and their ability to generate ATP. In in vivo studies, renal cortical l8 mitochondria were isolated from rabbits which received a toxic dose of cephaloridine (200 mg/kg) subcutaneously 2 hr before sacrifice. Ig_vjvg. exposure to cephaloridine resulted in a significant reduction of mito- chondrial respiratory rates. In jn_vitro studies, measurements of mitochondrial respiration were made before and after addition of 2000 ug/ml of cephaloridine, which was equal to the concentration found in renal cortex in which significant tubular necrosis was observed. This concentration reduced state 3 rates and RCR's with the natural sub- strates and respiration with tetramethylphenylenediamine plus ascorbate. The effect of cephaloridine on respiration was further examined in studies of isolated renal cortical tubules under three conditions: (a) ig_vjtrg_short-term exposure (5 to 10 min) to 2000 ug/ml of cephalori- dine; (b) jn_vjtrg_longer term exposure (2 hr) to the same concentra- tion; and (c) in vivo exposure by the subcutaneous injection of 400 mg/kg of cephaloridine 2 hours before the animals were sacrificed. Respiration rate was significantly reduced under the last two conditions (65 and 53% of control, respectively) but was not altered in the jg_ XiEEQ short-term exposure. One possible reason for this delayed expression of cephaloridine- induced cytotoxicity (inhibition of respiration by isolated renal tubules) is the requirement for biotransformation of cephaloridine to a toxic metabolite. Such a requirement was suggested by McMurtry and Mitchell (1977) and Mitchell et_gl, (1977). These investigators re- ported that two inhibitors of cytochrome P450 monooxygenase activities, cobaltous chloride and piperonyl butoxide, either completely blocked or l9 markedly reduced the renal proximal tubular necrosis produced by cepha- loridine in rats and mice (McMurtry and Mitchell, 1977). Furthermore, the lack of effects of these inhibitors on the direct nephrotoxicants, mercuric chloride and sodium fluoride, provided additional support for their hypothesis. They failed to increase cephaloridine nephrotoxicity by pretreatment of animals (rats and mice) with phenobarbital (McMurtry and Mitchell, 1977). However, phenobarbital has been shown to induce drug-metabolizing enzymes in the liver but not in the kidney in rats (Lake et_al,, 1973). The lack of effect of phenobarbital on renal drug- metabolizing enzyme activities as well as the lack of potentiation of cephaloridine toxicity by phenobarbital led these investigators (Mit- chell et_al,, 1977) to conclude that metabolic activation of the anti- biotic presumably occurred in situ in the kidney. Furthermore, these investigators reported that pretreatment of rats and mice with cysteine decreased cephaloridine renal injury and produced a shift in the zone of necrosis from the outer to the inner cortex. Even though the nature of the toxic cephaloridine metabolite was not demonstrated, on the basis of their findings, Mitchell et_al, (1977) suggested that it was an electro- philic reactant since nucleophilic sulfhydryl compounds decreased its toxicity. Furthermore, cephaloridine contains a 5-membered thiophene ring, and many thiophenes produced renal or hepatic necrosis in animals after metabolic activation (McMurtry and Mitchell, 1977), suggesting that a thiophene epoxide metabolite of cephaloridine might be generated. 20 Renal Drug-Metabolizing Enzymes Although the precise biochemical mechanism responsible for cepha- loridine nephrotoxicity is not clear, understanding of renal drug- metabolizing enzyme systems may help to extend knowledge of cephalori- dine toxicity. The kidney is one of the most sensitive organs in the body to the harmful effects of toxic chemicals. This high suscepti- bility to toxicants is the result of several factors. The two kidneys together receive approximately 25% of cardiac output, even though they comprise less than 1% of total body mass. High renal blood flow can rapidly deliver toxic chemicals to the kidneys. Because a large amount of plasma water is filtered and reabsorbed, and a variety of materials (e.g., cephaloridine) are actively transported into the tubular cells, many toxic chemicals can be concentrated in the kidney and have a greater opportunity to injure kidney cells than most other cells in the body. However, high concentrations of chemicals (e.g., penicillin G) do not necessarily produce cell damage. Under many conditions, chemicals (e.g., chloroform, acetaminophen and bromobenzene) have to be meta- bolically activated prior to causing cell injury. Since cephaloridine has been suggested to be bioactivated jn_§jtu_in the kidney, it becomes important to understand the characteristics of renal drug-metabolizing enzymes (DMEs). Many DME activities including cytochrome P450 monooxygenase, glu- curonyltransferase (Fowler gt_al,, 1977), glutathione S-transferase (Fine et_al,, 1978), epoxide hydrolase (Oesch, 1972) and others (Zenser gt_§l,, 1978; Orrenius et 31., 1973) are measurable in renal tissues. 21 However, the distribution of DMEs in the kidney appears to be hetero- geneous. Zenser et_al, (1978) found high activities of laureate hy- droxylase and aminopyrine demethylase in kidney cortex but little or no activity in medulla. Fine et 31. (1978) measured glutathione S-trans- ferase activity in isolated segments of the rabbit nephron using 1- chloro-2,4-dinitrobenzene as substrate. Enzyme activity was confined to the proximal tubules and was not detectable in the loop of Henle and collecting tubules. Tetrachlorodibenzo-p-dioxin (TCDD) has been admini- stered orally to rats and produced marked proliferation of smooth endo- plasmic reticulum in the renal proximal tubular cells of the straight segments. These changes were associated with pronounced induction of the microsomal enzymes glucuronyl transferase and benzo(a)pyrene hy- droxylase (Fowler et al,, 1977). Renal dissection studies disclosed that the activities of these enzymes in TCDD-treated animals were not uniformly distributed within the kidney. Enzyme activities were higher in the outer stripe of the medulla and cortex and lower in the medulla (Fowler et_al,, 1977). Therefore, in general, DME, especially mixed- function oxidases, are concentrated in the cells of the renal cortex. In general, renal enzyme activities appear to be lower than acti- vities of hepatic enzymes (Orrenius gt al., 1973). Based on a compara- tive study of drug metabolism by hepatic and extrahepatic tissue from five different species, Litterst gt al. (1975) reported that kidney activities were usually 15-40% of those found in liver. However, sub- strate specificity may be different for hepatic and extrahepatic enzymes. The renal MFO system has higher laurate hydroxylase activity 22 compared with the hepatic system in adult rabbits, suggesting that the kidney enzyme may metabolize fatty acid more rapidly (Zenser et 91-, 1978). In addition to substrate specificity, kidney tissues in several species of animals responded to barbiturate-type inducer (e.g., pheno- barbital and DDT) differently than did hepatic tissues. Phenobarbital increased hepatic but not renal microsomal enzyme activities in mice (Kluwe et_al,, 1978) and rats (Uehleke and Greim, 1968; Feuer et_al,, 1971). However, the lack of effect of phenobarbital on renal enzymes is not a uniform phenomenon. Phenobarbital was reported to induce MFO activities, cytochrome P450 and cytochrome b5 content in rabbit kidneys (Uehleke and Greim, 1968). Rabbits have been reported to be more susceptible to cephaloridine nephrotoxicity than rats (Atkinson §t_al,, 1966). This species difference in susceptibility may be due to quali- tative or quantitative differences in the ability of kidney cells to metabolize cephaloridine to reactive intermediates. As described previously, rabbit renal MFO activity can be induced by phenobarbital but rat and mouse MFO activities are not altered by phenobarbital treatment. This differential response suggests that the rabbit renal MFO system is somehow different from the rat and mouse MFO system. If this assumption is correct, then the rabbit kidney could have higher capacity to activate cephaloridine than rat and mouse kidney and there- fore produce greater damage in the rabbit kidney. Role of Glutathione in Cellular Protection Against Toxic Chemicals Glutathione conjugation has been considered a detoxification path- way for many toxic chemicals. Several metabolite-mediated toxic chemi- cals such as chloroform (Brown gt_al., 1974; Docks and Kirshna, 1976), 23 acetaminophen (McMurtry et al,, 1978) and vinyl chloride (Hefner gt 31., 1975) produce liver and/or kidney necrosis associated with depletion of glutathione (GSH). Cell injury or death results only under conditions in which this sulfhydryl compound is depleted (Mitchell et_al,, 1976). As described previously, Mitchell §t_al, (1977) reported that pretreat- ment of rats and mice with cysteine decreased cephaloridine renal injury. Since cysteine has been shown to be a precursor for glutathione synthesis (Meister and Tate, 1976; Boyland and Chasseaud, 1967; Reed and Orrenius, 1977), this finding suggests that cephaloridine or its meta- bolite(s) may react with glutathione. Glutathione (GSH) is a physio- logically important tripeptide with the sequence L-v-glutamyl-L-cys- tinylglycine. This ubiquitous molecule is the most abundant cellular peptide and accounts for about 90 percent of intracellular non-protein thiols. The structure of this tripeptide is unusual in that the amino- terminal peptide bond utilizes the v-carboxyl moiety of glutamate. This unusual arrangement makes GSH resistant to the hydrolytic action of many proteases and aminopeptidases. The cellular concentration of GSH at any given time is the result of a dynamic system and reflects the balance of degradation and biosynthesis processes. GSH is synthesized from its constituent amino acids (glutamate, cysteine and glycine) in two sepa- rate reactions, each requiring one molecule of ATP (Snoke and Bloch, 1952; Snoke gt 31., 1953). The first step involves formation of a v- glutamyl linkage between L—glutamate and L-cysteine, catalyzed by v- glutamyl cysteine synthetase (Meister, 1974). This reaction is rela- tively specific for L-glutamate but less specific for L-cysteine. In the second step, L-glycine is added to v-glutamyl cysteine to form GSH, 24 the reaction being catalyzed by glutathione synthetase (Meister, 1974). Regulation of GSH biosynthesis can be through feedback inhibition; GSH at physiological concentrations was reported to inhibit the formation of y-glutamylcysteine (Richman and Meister, 1975). Similarly, ADP was shown to inhibit glutathione synthetase (Snoke et 31., 1953; Yanari et .11., 1953) and v-glutamylcysteine synthetase (Mandeles and Block, 1955). In addition, the rate of GSH biosynthesis is limited by the availability of substrates (Tateishi et_al,, 1974). The breakdown of glutathione is catalyzed by y-glutamyl transpeptidase, a membrane-bound enzyme (Meister and Tate, 1976), which may be the only enzyme that is capable of hydro- lyzing the y-glutamyl-peptide bond since the usual peptidases appear to be unable to hydrolyze this linkage. The reaction catalyzed by y- glutamyltranspeptidase involves transfer of the v-glutamyl moiety of glutathione to an amino acid acceptor to yield a v-glutamyl peptide and cysteinylglycine. Cysteinylglycine is then split to cysteine and glycine by a peptidase (Semenaz, 1957; Hughey gt_al:, 1978). The most widely known biological role of GSH is to form GSH con- jugates with foreign compounds or their metabolites. These conjugation reactions occur non-enzymatically and are also catalyzed by glutathione S-transferases (Chasseaud, 1973). The glutathione S-conjugates are then converted to mercapturic acid through three sequential reactions. The first reaction is to remove the v-glutamyl moiety from the GSH con- jugates, which is catalyzed by y-glutamyl transpeptidase (Meister and Tate, 1976). The resulting S-conjugates of cysteinylglycine are then converted to the cysteine conjugates by removing glycine, the reaction 25 being catalyzed by a number of aminopeptidases and dipeptidases (Hughey §t_al,, 1978). The cysteine conjugates are subsequently acetylated by N-acetyltransferase to mercapturic acids (Green and Elce, 1975), which are nontoxic and quickly excreted. Conjugation with GSH not only faci- litates the excretion of toxic chemicals, but also serves to intercept highly reactive compounds before they can covalently bind to tissue macromolecules leading to cell damage or death. The role of GSH in detoxification of reactive metabolites has been studied with many com- pounds such as acetaminophen and bromobenzene. Evidence from the studies by Zampaglione et_al, (1973) and Jollow et al. (1974) demon- strated in vivo direct correlation between hepatic GSH concentrations and the severity of hepatic necrosis, the degree of covalent binding of bromobenzene to liver macromolecules and the urinary excretion of bromophenyl mercapturic acid. In addition, GSH has also been shown to exhibit a protective effect against acetaminophen-induced necrosis in the liver (Mitchell et_a1,, 1973) and kidney (McMurtry et_al,, 1978). In addition to the mercapturic acid detoxication pathway, via glutathione peroxidase glutathione can prevent cellular damage by blocking intracellular oxidative processes such as lipid peroxidation. Glutathione peroxidase, an enzyme found in most mammalian tissues, specifically uses GSH as a hydrOgen donor to reduce chemically reactive hydroperoxides (e.g., H202 and lipid hydroperoxides) to H20 or chemi- cally stable alcohols and thereby protect the integrity of the cells. Meanwhile two molecules of GSH are oxidized to GSSG (Cohen and Hoch- stein, 1963; Wendel, 1980) (ROOH + ZGSH + ROH + H20 + GSSG). The 26 resulting GSSG is then reduced back to GSH by glutathione reductase with NADPH, which is generated from the hexose monophosphate pathway (Beut- 1er, 1974) (GSSG + NADPH + H+ + ZGSH + NADP). Several lines of evidence demonstrated the role of glutathione peroxidase in preventing tissue lipid peroxidation. In jg_vitro studies, glutathione peroxidase was shown to prevent lipid peroxidation and functional impairment of iso- lated mitochondria (Flohe and Zimmermann, 1970). Furthermore, a gene- tically-determined deficiency Of glutathione peroxidase in human ery- throcytes has been correlated to drug-induced hemolysis and chronic hemolytic anemia (Necheles et al,, 1970; Steinberg and Necheles, 1971) when exposed to prooxidative drugs. In addition, selenium was shown to be an essential cofactor for glutathione peroxidase (Rotruck et 21,, 1973; Flohe gt_al,, 1973; Oh et 31., 1974) and animals fed with se- lenium-deficient diet had a decreased glutathione peroxidase activity (Rotruck gt_al,, 1973). The selenium-deficient animals were shown to be unable to prevent hydrogen peroxide-induced erythrocyte hemolysis (Rotruck et 21,, 1973) and became more susceptible to the toxicity of paraquat (Bus §t_al,, 1975), which was shown to act through generation of superoxide with subsequent initiation of lipid peroxidation (Bus at 31,. 1974; 1976). Lipid Peroxidation and Its Role in Cellular Toxicity Lipid peroxidation has been considered as a basic deteriorative reaction in the toxicity of a variety of xenobiotics (Plaa and Witschi, 1976). The peroxidative process is the reaction of oxidative deteriora- tion of polyunsaturated lipids. Peroxidation of lipids involves the 27 reaction of oxygen with polyunsaturated lipids to form free radical intermediates and lipid hydroperoxides, which then promote free radical chain oxidation (Tappel, 1973). Lipid peroxidation is damaging because of the subsequent reactions of free radicals, mainly peroxy radicals. Several endogenous oxidation reactions and oxidation of xenobiotics can convert oxygen to superoxide and hydroxyl radical. These various forms of oxygen are then capable of either directly or indirectly initiating lipid peroxidation (Tien et_al., 1981; Svingen gt_al,, 1978; Tyler, 1975; Fong gt al,, 1973; Kellogg and Fridovich, 1975, 1977). Generation of superoxide occurs in many endogenous biological reactions such as autooxidations of reduced flavins, hydroquinones and catecholamines and from the aerobic actions of enzymes such as xanthine oxidase, aldehyde oxidase and flavin dehydrogenases (McCay and Poyer, 1976; Fridovich, 1975). In addition, the toxicity of several chemicals such as adria- mycin (Goodman and Hochstein, 1977; Myers at al., 1977), alloxan (Heik- kila and Cohen, 1975), dialuric acid (Cohen and Heikkila, 1974), 6- hydroxydopamine (Heikkila and Cohen, 1973) and paraquat (Bus gt 21,, 1975) have been shown to be associated with superoxide formation. Plasma membranes and membranes of subcellular organelles are major sites of lipid peroxidative damage. Membranes are complex mixtures of lipids and proteins. Mitochondrial and microsomal membranes contain relatively large amounts of polyunsaturated fatty acids in their phos- pholipids (Rouser et 31., 1968). Furthermore, some of the most powerful catalysts involved in lipid peroxidation, such as coordinated iron and hemoproteins, are in close molecular proximity to these polyunsaturated 28 lipids (Tappel, 1973). Because of these particular characteristics, mitochondrial and microsomal membranes are highly susceptible to lipid peroxidative damage. Lipid peroxidative damage of mitochondrial mem- branes has been demonstrated to correlate with swelling and lysis of the mitochondria and impaired mitochondrial respiration (Hofsten et_al,, 1962; Vladimirov et_al., 1980). Lipid peroxidation also affects micro- somal membranes, decreasing microsomal drug metabolism and cytochrome P450 concentration (Plaa and Witschi, 1976). In addition, lipid peroxi- dation damages lysosomal membranes (Fong gt_al,, 1973; Wills and Wilkin- son, 1966), with resultant intracellular release of hydrolytic enzymes. Proteins and enzymes in aqueous solution are also susceptible to lipid peroxidative damage (Tappel, 1973). Thus, a great number of cellular functions are altered as a result of lipid peroxidation (Vladimirov gt. 21,, 1980). Since lipid peroxidation is an exceedingly damaging biological process and the peroxidative process is very likely constantly occurring in the cells, in order to survive cells possess several protective mechanisms. Superoxide dismutase has been shown to scavenge the super- oxide radicals, which will prevent the initiation of lipid peroxidation (Hassan and Fridovich, 1980). Several endogenous antioxidants such as glutathione and vitamin E are known to prevent the propagation of lipid peroxidation (Tappel, 1978). The roles of glutathione, glutathione peroxidase and selenium against lipid peroxidation have been described previously. Vitamin E is an important nutritional antioxidant and a free radical scavenger (Wasserman and Taylor, 1972; Tappel and Green, 1972; Urano and Matsui, 1976), which is able to interrupt the free 29 radical chain reactions of lipid peroxidation and prevent peroxidative damage (Tappel, 1978). Tappel and Zalkin (1959) and Dillard and Tappel (1971) have shown that liver mitochondria and microsomes isolated from rats fed vitamin E-deficient diets had greater peroxidation rates than rats fed basal diets supplemented with vitamin E. Expiration of pentane and ethane deriving from lipid peroxidation (Dumelin and Tappel, 1977), has been used as an jn_vjv9_index of lipid peroxidation (Riely et 31., 1974; Dillard gt_al,, 1977). Dillard et_al, (1978) demonstrated that dietary vitamin E decreased pentane expiration in rats. Hafeman and Hoekstra (1977) also reported ethane expiration in carbon tetrachloride- treated rats to be diminished by dietary vitamin E. Vitamin E defi- ciency was reported to increase the susceptibility of animals to the toxicity of oxygen (Mino, 1973), ozone (Goldstein et 21,, 1970) and paraquat (Bus et_al., 1975), which were shown to be associated with lipid peroxidative damage. Tudhope and Hopkins (1975) also reported an increased susceptibility of erythrocytes to lipid peroxidation on expo- sure to hydrogen peroxide vapor when erythrocytes were obtained from patients with low plasma vitamin E. Thus, by altering vitamin E (anti- oxidant) and selenium (a cofactor of glutathione peroxidase) concentra- tions in the diet, it has become possible to evaluate indirectly lipid peroxidation jn_vjvg, One of the common features between paraquat (Figure 2) and cepha- loridine (Figure 1) is the pyridinium ring. Paraquat contains two pyridinium rings and cephaloridine has one pyridinium substitute on the B-lactam ring. Pyridinyl free radicals have been found to exist in a 30 .uescmcea mo weapozcum one .m mcamwm 31 variety of pyridinium-containing compounds (Kosower, 1976). Studies by Mees (1960), Gage (1968), Farrington (1973), Davies and Davies (1974) and Bus at 31, (1975, 1976) demonstrated that paraquat can be reduced to the paraquat pyridinyl free radical by a single electron reduction reaction catalyzed by cytochrome P450 reductase and NADPH (Pederson and Aust, 1973; Pederson et_al,, 1973); Paraquat free radical is then non- enzymatically converted back to paraquat and meanwhile superoxide radicals are generated from oxygen (Figure 3). Superoxide radicals can subsequently initiate reactions with unsaturated lipids associated with cell membranes to form lipid hydroperoxides. The pyridinium ring of cephaloridine may be metabolized through this reduction-oxidation cycle and produce superoxide radicals (Figure 4). Objectives The purposes of this investigation were three-fold: (a) to deter- mine the role of biotransformation in cephaloridine-induced nephro- toxicity; (b) to evaluate the role of glutathione conjugation against cephaloridine toxicity; and (c) to estimate the role of lipid peroxi- dation in cephaloridine nephrotoxicity. 32 .mc_xocwq:m mo cowpeELoe pemuwsoucou use pescegea mo cowpeuwxo-compu=umc UPFUao one .m mezmwu m I l I In .mo 5 age :uJ m0<2 Imo<2 33 -> R — NO— \ < NADPH O._> NADP _©\> <02 02 — LIPID PEROXIDATION l2 .__ Figure 4. Proposed cyclic reduction and oxidation of cephaloridine. METHODS Animals Male Fischer 344 rats (200-250 g) and female New Zealand White rabbits (1.5-3.0 kg) were purchased from Harlan Industries, Inc. (In- dianapolis, IN) and a local breeder, respectively. In a few experi- ments, male Sprague-Dawley (SD) rats (200-250 g) and male ICR mice (25- 35 9) obtained from Spartan Farms (Haslett, MI), were also used. Animals were maintained under standardized conditions of light (7 a.m.-7 p.m.) and temperature (25:2°C) and allowed free access to food (Wayne Lab-Blox, Allied Mills, Inc., Chicago, IL; Lab Rabbit Chow No. 5321, Ralston Purina Co., St. Louis, MO) and water until use. In the dietary study, 5-6 week old male Fischer 344 rats (56-85 g) were purhcased from Harlan Industries, Inc. and housed in wire mesh hanging cages, maintained in a room with a 12 hr light-12 hr dark cycle and allowed free access to food and water. Induction of Renal Microsomal Monoxygenases and Glutathione S-Trans- ferases by Phenobarbital Rats and rabbits were given a single i.p. injection of phenobar- bital (80 and 60 mg/kg, respectively) once daily for four days. Animals were killed by cervical dislocation 24 hr after the last dose. Livers and kidneys were quickly excised and placed in ice-cold 1.15% KCl. 34 35 After being weighed, renal medulla and papilla were carefully dissected out and the remaining cortical portion was minced in 20 mM Tris-HCl buffer (pH 7.4) containing 1.15% KCl, rinsed 3 times and homogenized in 3 volumes of the same solution using a Potter-Elvehjem homogenizer with a Teflon pestle followed by centrifugation at 10,000 x g for 20 minutes. The resulting supernatant was then centrifuged at 100,000 x g for 60 minutes. The pellet was resuspended in 66 mM Tris-HCl buffer, pH 7.4 and recentrifuged at 100,000 x g for 60 minutes. The pellet was re- suspended in 66 mM Tris-HCl buffer, pH 7.4 containing 0.25 M sucrose and 5.4 mM EDTA to a final concentration of 10-20 mg protein/ml. The activities of benzphetamine-N-demethylase (Prough and Ziegler, 1977), and ethoxycoumarin-O-deethylase (Ullrich and Weber, 1972) were then determined in the final microsomal suspension. Cytochrome P450 concentrations were determined according to the method of Omura and Sato (1964). Protein was quantified by the method of Lowry (1951). In the second series of experiments, animals were treated similarly and after cervical dislocation, portions of liver and renal cortex were minced and homogenized in 3 volumes of ice-cold 66 mM Tris-HCl buffer, pH 7.4. The homogenates were centrifuged at 10,000 x g for 20 minutes. The resulting supernatants were further centrifuged at 100,000 x g for 60 min. Arylhydrocarbon hydroxylase activity was measured in the 10,000 x g supernatant by the method of Nebert and Gelboin (1968) as modified by Oesch (1976) with quinine sulfate as the standard. Gluta- thione S-transferase activity was measured in the 100,000 x g superna- tant using 1-ch1oro-2,4-dinitrobenzene as the substrate (Habig gt al., 1974). 36 Effects of Phenobarbital Pretreatment on Cephaloridine Toxicity Rabbits and rats received 60 and 80 mg/kg of phenobarbital in saline, i.p., respectively, once daily for 4 days. The control group received saline alone. Twenty-four hours after the last injection of phenobarbital or saline, rabbits and rats were dosed with a single administration of cephaloridine (Eli Lilly and Company, Indianapolis, IN). Rabbits were killed 48 hr later. Because in the preliminary experiments, some rats died between 24 and 48 hr after administration of cephaloridine, all rats were killed 24 hr following cephaloridine administration. Nephrotoxicity was evaluated by determining changes in blood urea nitrogen (BUN) and p-aminohippurate (PAH) and tetraethyl- ammonium (TEA) accumulation in renal cortical slices. After animals were killed, blood samples were collected for determination of BUN and serum glutamic pyruvic transaminase activity (SGPT). Livers and kidneys were removed immediately and weighed. Renal cortical slices were pre- pared for determination of PAH and TEA accumulation. Effects of Piperonyl Butoxide Pretreatment on Cephaloridine Toxicity Rats received a single i.p. injection of piperonyl butoxide (1000 mg/kg) 30 min prior to a single i.p. administration of cephaloridine. The control animals received corn oil prior to cephaloridine admini- stration. All animals were killed 24 hr after cephaloridine admini- stration. BUN and renal cortical accumulation of PAH and TEA were determined. In the second series of experiments, rabbits received a single i.p. injection of piperonyl butoxide (135 or 750 mg/kg) 30 min prior to a single s.c. administration of cephaloridine (150 mg/kg) and 37 were killed 48 hours later. Histopathological alterations in renal cortical tissues were determined. Effects of Phenobarbital Pretreatment on Renal Cortical Uptake and Runout of Cephaloridine In the first series of experiments, animals were pretreated with phenobarbital or saline as described previously, and killed one, two or three hours after a single administration of cephaloridine (150 or 1000 mg/kg). After animals were killed, blood was collected to prepare serum samples, tissues were removed immediately and portions of the tissues were homogenized in five volumes of distilled water using a Potter- Elvehjem homogenizer. Concentrations of cephaloridine in sera and tissue homogenates were then determined. In the second series of experiments, rabbits were treated with phenobarbital or saline as described previously. Twenty-four hours after the last phenobarbital injections, animals were killed and renal cortical slices were prepared. Cortical slices were then incubated in 4.0 ml of phosphate—buffered medium containing 275 pg/ml of cephalori- dine at 25°C under 100% 02 for 45 and 90 minutes. Following incubation, slices were removed, blotted, weighed and homogenized in one ml of water and mixed with one ml of CH3CN. In addition, one ml of medium was mixed with one m1 of CH3CN. Concentrations of cephaloridine in these mix- tures were then determined. Cephaloridine runout was determined by the method described by Farah et_al, (1963) with a few modifications. Kidney slices were preloaded with cephaloridine by incubating tissue in 4.0 ml of medium 38 which contained 250 pg/ml of cephaloridine under 100% 02, at 25°C for 90 min. Slices were then removed, rinsed and transferred at l min inter- vals through a series of beakers containing 4.0 ml of cephaloridine-free medium. At 0, 5, 10, 15, 20 and 25 min, slices were removed, blotted, weighed, homogenized in 1 m1 of H20 and mixed with 1.0 m1 of CH3CN. Cephaloridine concentration in the homogenates was then determined. Effects of Piperonyl Butoxide Pretreatment on Renal Cortical Accumula- tion of Cephaloridine In the first series of experiments, rabbits received a single i.p. injection of piperonyl butoxide 30 min prior to administration of cephaloridine (150 mg/kg) and were killed 15 or 60 min after cepha- loridine administration. Serum and renal cortical concentrations of cephaloridine were determined spectrophotometrically. In the second series of experiments, rabbits received a single i.p. injection of piperonyl butoxide (750 mg/kg) or 1% Tween 80 (vehicle) and were killed 45 min later. Cortical slices were prepared and incubated in 4.0 m1 of phosphate-buffered medium containing 125 ug/ml of cepha- loridine at 25°C under 100% 02 for 30, 60 and 90 minutes. Following incubation, slices and medium were treated as described previously and cephaloridine concentrations in the slices and medium were then deter- mined. Effects of Phenobarbital Pretreatment on Renal Cortical Accumulation of Inulin and p-Aminohippurate In Vivo and Cortical Slice Accumfilation of p-Aminohippurate and Tetraethylammonium In Vitro Rabbits were pretreated with phenobarbital or saline as described previously. Twenty-four hours after the last phenobarbital injection, 39 animals were infused with a solution containing PAH and inulin according to the method of Tune and Fernholt (1973). Rabbits were given a priming dose of PAH, 17.5 mg/kg body weight, and 2.0 ml of 10% inulin solution through an ear vein. A solution containing 10% inulin and 20 mg/ml of PAH was then infused, i.v., at a rate of 0.11 ml/min. After one hr of infusion the animals were killed by cervical dislocation. Blood was collected and allowed to clot for one hr at room temperature and serum was prepared by centrifugation at 2,000 x g for 10 min. A 0.25 ml aliquot of serum was mixed with 0.6 ml 10% TCA, brought to a final volume of 2.0 ml with distilled water and centrifuged at 2,000 x g for 10 min and the precipitate discarded. Portions of renal cortex and liver were homogenized with 20 volumes of 3% TCA and centrifuged at 2,000 x g for 10 min. PAH concentrations in the resulting supernatants from tissue and serum samples were then determined by the method of Smith gt_al, (1945) and inulin concentrations in the tissue supernatants were also measured following the method of Schreiner (1950). In addi— tion, a 0.1 ml aliquot of serum was mixed with 0.2 ml of 0.75 N NaOH, 0.2 ml of ZnSO4 in H2504 (100 g of ZnSO4-7H20 and 40 m1 of 6N H2504 diluted to 1000 ml with distilled water) and 1.5 ml distilled water and centrifuged at 2,000 x g for 20 min. Inulin concentrations in the supernatants were then determined by the method of Walser gt_al, (1955). In the second series of experiments, rabbits were pretreated with phenobarbital or saline four days. Twenty-four hours after the last phenobarbital or saline injection, animals were given a bolus ear vein injection of PAH (40 mg/kg) and inulin (200 mg/kg) in saline and killed 40 one hour later. Serum and tissue PAH and inulin concentrations were determined as described previously. In the third series of experiments, rabbits were treated with phenobarbital or saline as described previously. Twenty-four hours after the last phenobarbital injections, animals were killed and renal cortical slices were prepared. Accumulation of PAH and TEA in renal cortical slices were then measured. Effects of Piperonyl Butoxide Pretreatment on Renal Cortical Accumula- tion of p-Aminohippurate, Tetraethylammonium and Inulin In the first series of experiments, rabbits received a single i.p. injection of piperonyl butoxide (750 mg/kg) 30 or 90 min prior to in- fusion with a solution containing PAH and inulin according to the method of Tune and Fernholt (1973) as described previously. In the second series of experiments, animals received piperonyl butoxide (750 mg/kg) in 1% Tween 80 (vehicle). Control animals received the vehicle alone. Forty-five minutes later animals were killed and uptake of PAH and TEA by renal cortical slices were determined. Depletion of Renal Glutathione Concentration and Nephrotoxicity of Cephaloridine in Rabbits, Rats and Mice The purpose of the first series of experiments was to determine cephaloridine toxicity in rabbits, rats and mice forty-eight hours following cephaloridine administration. Animals were administered cephaloridine i.p. or s.c., at doses ranging from 150 to 2000 mg/kg in saline. Control animals were given saline only. All animals were killed 48 hours later by cervical dislocation followed by decapitation. 41 BUN and SGPT activity in the serum were measured. Livers and kidneys were removed immediately and weighed. PAH and TEA accumulation by renal cortical slices were determined. The second series of experiments was designed to quantify gluta- thione (GSH) depletion in kidneys and livers one to four hr following a single administration of cephaloridine. In order to avoid diurnal changes in tissue GSH concentration (Jaeger gt gl., 1973; Hassing gt gl,, 1979), animals were dosed between 7 a.m. and 9 a.m. and killed before 1 p.m. GSH and water content in the livers and kidneys were determined. In addition, PAH and TEA accumulation and gluconeogenesis in renal cortical slices were measured. In the third series of experiments, animals were treated with diethyl maleate (0.4 ml/kg, i.p.) in corn oil, 30 min prior to admini- stration of cephaloridine. Animals were killed 24 or 48 hours later and PAH and TEA accumulation in renal cortical slices were determined. Blood urea nitrogen was also measured. Dose-Dependent Effect of Cephaloridine, Cephalothin and Gentamicin on Tissue GSH Concentration Experiments were designed to quantify GSH depletion in tissues shortly following a single administration of cephaloridine, cephalothin (Eli Lilly and Company, Indianapolis, IN) or gentamicin (Schering Corporation, Kenilworth, NJ). Animals were dosed between 7 a.m. and 9 a.m. and killed before 10 a.m. One hour after a single i.p. injection of cephaloridine (0-2000 mg/kg) or cephalothin (0-2000 mg/kg), animals were killed, kidneys and livers immediately removed and tissue GSH was 42 measured. In addition, some animals received a single i.p. injection of gentamicin (O-lOOO mg/kg). Because some animals receiving the highest dose of gentamicin died within 45 min, all animals were killed at 45 min instead of 60 min. Effect of Cephaloridine on Tissue Reduced and Oxidized Glutathione (GSH and GSSG) and Conjugated Dienes In the first series of experiments, rabbits and rats received a single administration of cephaloridine or saline and were killed one hour later. Concentrations of reduced and oxidized glutathione in the liver and renal cortex were determined immediately. In the second series of experiments, rats and rabbits received a single administration of cephaloridine and were killed at various times. Portions of liver and renal cortex were used to prepare microsomes as described previously and concentrations of conjugated diene in the microsomes were then determined. Effect of Vitamin E and/or Selenium Deficiency on Cephaloridine Toxicity in Rats Five to six weeks old male Fischer 344 rats (56-85 g) were housed in wire mesh hanging cages. Diets (Teklad Test Diets, Madison WI) and water were supplied gg libitum. The tolura yeast-basal diet (Tables 1 and 2) was deficient in vitamin E and selenium (Se). Supplemental levels of 121 U/kg vitamin E and 0.2 ppm Se were used. Animals were divided into four groups. Group 1 was fed the basal diet with no additional vitamin E or Se, group 2 received vitamin E-supplemented diet, group 3 received 0.2 ppm Se-supplemented diet and group 4 received 43 TABLE 1 Composition of Basal Diet Ingredient Composition (g/kg) Torula Yeast 300.0 DL-Methionine 3.0 Sucrose 593.98 Lard, Tocopherol-Stripped 50.0 Mineral Mix, Hubbell-Mendel-Wakeman 50.0 Manganese Sulfate 0.154 p-Aminobenzoic Acid 0.11 Biotin 0.0004 Vitamin B12 (0.1% trituration in mannitol) 0.03 Calcium Pantothenate 0.04 Choline Dihydrogen Citrate 2.44 Folic Acid 0.002 Menandione 0.05 Niacin 0.05 Pyridoxine HCl 0.015 Riboflavin 0.01 Thiamin HCl 0.01 Vitamin A Palmitate, in Corn Oil (200,000 U/g) 0.099 Vitamin 02’ in Corn Oil (400,000 U/g) 0.0055 44 TABLE 2 Mineral Mix, Hubbell-Mendel-Wakeman Ingredient Composition (g/kg) Calcium Carbonate (CaCO3) 543.0 Magnesium Carbonate (MgC03) 25.0 Magnesium Sulfate (M9804) 16.0 Sodium Chloride (NaCl) 69.0 Potassium Chloride (KCl) 112.0 Potassium Phosphate, Monobasic (KH2P04) 212.0 Ferric Pyrophosphate 20.5 Potassium Iodide (KI) 0.08 Manganese Sulfate (MnSO4-H20) 0.35 Sodium Fluoride (NaF) 1.0 Aluminum Potassium Sulfate (A1K(SO4)2-12H20) 0.17 Cupric Sulfate (CuSO4) 0.90 45 both supplements. Forty—two days after being fed the experimental diets, animals received a single i.p. injection of cephaloridine (O, 500 or 1000 mg/kg) in saline and were killed 24 hr later. BUN and SGPT activity in the serum were measured. Kidneys and livers were removed and weighed. Gluconeogenesis by renal cortical slices were determined. HistopatholOgical alterations in the kidneys and livers were also examined. Analytical Methods Determination of PAH and TEA Accumulation and Gluconeogenesis in Renal Cortical Slices. After animals were killed, the kidneys were removed and placed in ice-cold saline until use. Thin renal cortical slices were prepared and incubated in 4.0 m1 of phosphate-buffered 5M PAH and medium (Cross and Taggart, 1950) which contained 7.4x10' 1.0x10'5M [14CJTEA (specific activity, 2.0 Ci/mole, New England Nuclear, Boston, MA). Incubations were carried out in a Dubnoff metabolic shaker at 25°C under a gas phase of 100% 02 for 90 min. After incubation, the slices were removed, weighed and homogenized in 10 ml of 3% trichloro- acetic acid. Two m1 of incubation medium was treated similarly. After centrifugation, the supernatant was assayed for PAH and [14CJTEA con- centrations. PAH was determined by the method of Smith gt_gl, (1945). [14CJTEA, one m1 of slice or medium supernatant was added to To quantify 10 m1 of ACS counting cocktail (Amersham, Arlington Heights, IL) and radioactivity was determined. The accumulation of PAH and TEA in renal cortical slices was expressed as a slice-to-medium (S/M) concentration 46 ratio, where S represents mg of PAH or TEA per gram of tissue and M represents mg of PAH or TEA per ml of medium. Gluconeogenesis by renal cortical slices jg_gittg_was measured according to the method of Roobol and Alleyne (1974) with a few modi- fications. Approximately 100 mg of tissue slices were incubated in 5.0 ml of Krebs-Ringer bicarbonate medium containing 10 mM pyruvate. The incubation flasks with the media were gassed with 95% 02/5% C02, capped tightly, and incubated for 60 min at 37°C in a Dubnoff metabolic shaker. The tissues were then removed, blotted and weighed. After removal of the tissue, the incubation medium was centrifuged at 2,000 x g for 10 min, and glucose concentration in the supernatant was determined using reagents obtained from Sigma Chemical Co. (Sigma Technical Bulletin No. 510, St. Louis, MO). Determination of Blood Urea Nitrogen (BUN) and Serum Glutamic Pyruvic Transaminase (SGPT) Activity. The blood samples were allowed to clot at room temperature and centrifuged at 600 x g for 10 min and BUN and SGPT activity in the sera were then determined. BUN was measured spectrophotometrically (Fawcett and Scott, 1960; Chaney and Marbach, 1962) with Sigma reagents (Sigma Technical Bulletin No. 640, Sigma Chemical Co., St. Louis, MO). SGPT activity was assayed (Reitman and Frankel, 1957) with Sigma reagents (Sigma Technical Bulletin No. 505) and the activity was expressed as Sigma-Frankel (SF) units/ml. One SF unit of SGPT activity will form 4.82x104 umoles glutamate/min in phosphate buffer, le7.5 at 25°C. 47 Determination of Tissue Water Content and Reduced Glutathione (GSH) and Oxidized Glutathione (GSSG) Concentration. Water content was determined as weight loss after drying to constant weight at 85°C. Reduced glutathione (GSH), more precisely non-protein sulfhydryl con- tent, was measured according to the method of Ellman (1959) with a few modifications. Kidney or liver portions were homogenized in 20 volumes of ice-cold % trichloroacetic acid (TCA) and centrifuged at 10,000 x g for 20 min. After an adequate dilution with ice-cold 6% TCA, 0.5 ml of the diluted supernatant was added to 2 ml of 0.3 M Na2HP04 solution. 0.5 m1 of 0.04% 5,5'-dithio-bis-(2-nitrobenzoic acid) in 10% sodium citrate was then added and immediately after mixing the absorbance at 412 nm was measured. In the second series of experiments, both reduced and oxidized glutathione (GSH and GSSG), more precisely both reduced and oxidized non-protein sulfhydryl, were determined according to the method of Ellman (1959) modified by Van Doorn gt_gl. (1978). After animals were killed, portions of liver and renal cortex (0.4-0.6 g) were homogenized in 20 volumes of ice-cold 0.15 M KCl containing 30 mM EDTA. For the determination of GSH, 2 ml of homogenates were deproteinized by the addition of 3 m1 of a solution containing 0.3 g/ml NaCl, 0.0167 g/ml metaphosphoric acid and 0.002 g/ml EDTA. After centrifugation at 10,000 x g for 20 min at 4°C, 0.5 m1 of the diluted supernatant was added to 2 m1 of 0.3 M NazHPO4 solution. 0.5 m1 of 0.04% 5,5'-dithio- bis-(Z—nitrobenzoic acid) in 10% sodium citrate was then added and the absorbance at 412 nm was measured immediately after mixing. Total glutathione (GSH and GSSG) was assayed as follows: One ml of 48 deproteinized supernatant was reduced with 1 m1 of 5% NaBH4 and then incubated at 45°C for 60 min. The mixture was neutralized with 0.5 m1 of 2.7 N HCl and SH groups were assayed as described previously. Measurement of Conjugated Diene in Microsomes. Conjugated diene concentrations of microsomal lipids were determined by the method of Recknagel and Ghoshal (1966) modified by Sell and Reynolds (1969). One to two grams of liver or renal cortex were homogenized in 3 volumes of ice-cold 0.3 M sucrose, 0.003 M EDTA solution and centrifuged at 10,000 x g for 20 min. The resulting supernatants were further centrifuged at 100,000 x g for 60 min. The pellets were weighed and an adequate volume of 0.3 M sucrose, 0.003 M EDTA solution was added to make a final concentration of 300 mg microsome/ml. 0.5 m1 of microsomal suspension was extracted with 9.5 m1 of chloroformzmethanol (2:1) solution. The mixture was shaken for 15 min and filtered. The filtrate was separated into two phases by the addition of 2.0 m1 of water and the upper phase was discarded. 0.2 m1 of methanol was added to the lower phase and the absorbance was determined between 210 and 300 nm using freshly prepared lower phase as blank (25 ml water to 95 m1 2:1 chloroformzmethanol). Peak absorption for diene conjugation products of lipid peroxidation was at 240 nm. Absorbance at 240 nm, expressed as 00 units per gram micro- somes, was used as an indicator of the extent of lipid peroxidation. Determination of Serum and Tissue Cephaloridine Concentration. After animals were killed, blood was collected to prepare serum samples. Tissue were removed immediately, and portions of the tissues were homogenized in five volumes of distilled water using a Potter- Elvehjem homogenizer. A 0.3 ml aliquot of the homogenate was mixed with 49 0.7 ml of acetonitrile (CH3CN). Meanwhile a 0.3 ml aliquot of serum was also mixed with 0.7 ml of CH3CN. The mixtures were centrifuged at 2,000 x g for 10 min and precipitates discarded. The supernatants were removed and filtered through a 0.45 p regenerated cellulose membrane (Anspec Inc., Ann Arbor, MI) and 10-50 pl aliquots were injected onto an HPLC column. HPLC analyses were performed on a Model 440 chromatograph (Waters Associates, Inc., Milford, MA), equipped with a model U6K injector (Waters Associates, Inc.) at ambient temperature. A Radial-PAK C18 reverse phase column (8 mm ID x 10 cm) was used. The mobile phase consisted of 6 % CH3CN/4O% H20 at a flow rate of 2.5 ml/min. Pressure was 1,000-l,500 p.s.i. Effluent absorbance was monitored at 254 nm and recorded at 0.02-0.l absorbance units full scale. Peak heights or areas were determined and converted to concentrations of cephaloridine by interpolation from a standard curve. In some experiments, cephaloridine concentrations in serum and tissue were determined according to the method of Tune (1972). After animals were killed, approximately 0.1 g renal cortex was homogenized in 5 m1 of distilled water. Serum or tissue suspensions were diluted with distilled water to achieve cephaloridine concentrations of 5 to 10 pg/ml. 500 pl of each solution were then added to 2 m1 of 8.75% tri- chloroacetic acid to precipitate protein. After centrifugation, 500 pl of the supernatant were added to 500 p1 of 2 N HCl and the mixture was heated in a stoppered tube at 98°C for one hour to release pyridine by acid hydrolysis. The mixture was then allowed to cool to room tempera- ture. One drop of 0.1% methyl red in 50% aqueous ethanol was added. The mixture was titrated to a yellow color (pH 6.2) with 2 N sodium 50 hydroxide and left at room temperature for three hours. To each tube were added 1 ml of a freshly prepared solution of 1% aniline in a 2.5% sodium phosphate buffer (pH 6.0) and 500 pl of fresh cyanogen bromide (1 m1 of bromine in 50 ml of distilled water, decolorized by dropwise addition of 10% potassium cyanide). Finally, water was added to adjust the total volume to 5 ml. Fluorsecence was read six minutes after the last addition with excitation at 475 nm and emission at 515 nm. Determination of Cephaloridine Concentrations in Renal Cortical Slices and Incubation Media. Following incubation, cortical slices were removed, blotted, weighed and homogenized in one m1 of water and mixed with one m1 of CH3CN. The homogenates were then centrifuged and filtered. In addition, one ml of medium was mixed with one m1 of CH3CN and treated similarly. A 10-50 pl aliquot of filtrate was injected onto the HPLC column. HPLC analyses were conducted in a similar manner as described previously with a few modifications. A 5 mm ID x 10 cm Radial-PAK C18 column was used. The mobile phase consisted of 3 % CH3CN/70% H20/5 mM PIC B8 (Waters Associates, Inc.). A flow rate of 2.5 m1/min was maintained, with a pressure of l,000-1,500 p.s.i. Determination of Potential GSH Conjugates of Cephaloridine or Its Metabolite(s). The purpose of the first series of experiments was to determine the possibility of nonenzymatically catalyzed reactions be- tween GSH and cephaloridine. A final volume of 100 pl incubation mixture containing 50 mM Tris-HCl (pH 7.2-9.0) or 50 mM sodium phosphate buffer (pH 5.7-7.5), 120.pg/pi cephaloridine and 10 mM [35SJGSH (speci- fic activity 6.9 Ci/mole, New England Nuclear, Boston, MA) was incubated 51 at 25° or 37°C for 60 min. The control blanks only contained 50 mM buffer and 10 mM [3SSJGSH. After incubation, samples were then analyzed by thin layer and high pressure liquid chromatography. In the second series of experiments, possible enzymatically cata— lyzed GSH conjugating reactions were investigated. A portion of non- treated rabbit or rat hepatic or cortical tissues was homogenized in 3 volumes of 66 mM Tris-HCL buffer, pH 7.4 or 50 mM sodium phosphate buffer, pH 7.0 and then centrifuged at 10,000 x g for 20 min. The resulting supernatant was then used as an enzyme source. In some experiments, 10,000 x g supernatant was further centrifuged at 100,000 x g for 60 min. The resulting 100,000 x g pellet was resuspended in either 66 mM Tris-HCl or 50 mM sodium phosphate buffer to a final con- centration of 10-20 mg/ml (microsomal fraction). A final volume of 250 p1 incubation mixture containing 10-50 pl enzymes (10,000 x 9 super- natant, 100,000 x g supernatant or microsomal fraction), 25 pg/pl cephaloridine, 5 mM [BSSJGSH (specific activity, 6.9 Ci/mole), 3 mM MgCl2 and 1 mM NADPH was incubated at 37°C for 30 min. The control b1ank(s) contained the same ingredients but the enzyme sources. The denatured enzymes (boiled at 100°C for 5 min) were used in the control blanks. After incubation, the samples were analyzed by TLC and HPLC. For TLC analysis, an aliquot (50-150 pl) of incubation mixture was spotted on an activated silica gel GF plate (Analtech, Inc., Newark, Delaware), dried under air flow, and then developed with a solvent system consisting of n-butanol:acetic acid:water (2:1:1, by volume). After being developed, every l-cm wide band was scrapped into a 52 scintillation vial containing 10 ml Aquasol-2 solution (New England Nuclear, Boston, MA) and 4 ml water. The samples were mixed well and radioactivities were determined. For HPLC analysis, after incubation, the incubation mixture was deproteinized by the addition of an equal volume of CH3CN and centri- fugation. The supernatant was filtered and an aliquot of filtrate (10- 100 p) was then injected onto a Radial Pak C18 reverse phase column (8 mm ID x 10 cm). The mobile phase consisted of CH3CN and water at a flow rate of 2.0 ml/min and a linear gradient from 0.5 to 60% CH3CN was used. Effluent absorbance was monitored at 254 nm and every 1.0 ml of effluent was collected in a scintillation vial containing 10 ml ACS and the radioactivity was determined. Histology After animals were killed, portions of kidney tissues were quickly fixed in 10% neutral buffered formalin (3.7% formaldehyde in 75 mM sodium phosphate buffer, pH 7.0). After fixation, tissue samples were embedded in paraffin. Sectionss of kidney were then stained with hema- toxylin and eosin (H and E), and evaluated for glomerular, tubular and papillary abnormalities. Degrees of renal cortical necrosis were quantified as follows: mild, less than 10% of proximal tubules necro- tic; moderate, 10 to less than 25% of tubules necrotic; severe, 25% or more of the tubules necrotic. 53 Statistics The data were analyzed by Student's t-test or analysis of variance, completely randomized design, and treatment means were compared with the least-significant difference test (Steel and Torrie, 1960). The 0.05 level of probability was used as the criterion of significance. RESULTS Induction of Renal Microsomal Monooxygenases and Glutathione S- Transferase Activities Treatment with phenobarbital for four consecutive days had no effect on rat renal cytochrome P450 concentration, ethoxycoumarin-O- deethylase or arylhydrocarbon hydroxylase activity. In contrast, phenobarbital treatment increased rabbit renal cytochrome P450 content (385% of control), ethoxycoumarin-O-deethylase (140 %), benzphetamine- N-demethylase (422%) and arylhydrocarbon hydroxylase (440%) activities (Table 3). In control animals, both renal and hepatic glutathione S-trans- ferase activities were higher in rabbits than in rats (Table 4); rabbit renal enzyme activity was approximately 6-fold higher than rat. Treat- ment with phenobarbital significantly increased renal enzyme activity in rabbits but not in rats (Table 4). In contrast, the same treatment increased rat hepatic enzyme activity but had little effect on rabbit hepatic enzyme activity. Inhibition of Renal Glutathione S-Transferase Activities by Cephaloridine In Vitro Cephaloridine inhibited renal glutathione S-transferase activities in a dose-dependent manner. For example, at concentrations of 1.8 and 54 .Amo.oVav m—ocpcou Eoce ucmgmemwu Apuceuwwwcmwme .cwe\cwmpocn me\muP:= accommco=Fm m>_pepmmm .cws\cwmuoga ms\umELom mnxsmupescow mmPoEcc .cws\cwmuoga ms\nmEcoe secessooxxocnagim mmposao .cwmuoga Pmeomocows ae\mmpo5cn .mFeEwcm egos so Lao; mo mm a memos wee mmzpm> .mcoFm mcw_mm um>wmumc mFeswce Focucou .compumwcw umep Levee e; um cmppwx mew: use mzmn Lace com waec muco .zpm>wp -umgmmc ._epwnceao:m;q mo mx\me ow use om mo cowuomncw a? mrmcwm e um>wmumc mum; uce mpmnneme 55 mm._nmm.e oe.onm_.m com.__neK.MN mm._nme.e ceme_axecezz eeneeeeee»;Fac< --- --- cm.Fno.Kmm o.eno.emw acme.»eeeeeo-z eeeEeeeeeneem N._ne.e_ a.mnm.e_ cp.menm.mmm_ o.emnm.em eema_»eeeeo-o ewcee=eexxeepm eoo.onmmo.o ~oo.oneeo.o eomo.onmme.o mmo.on~_P.o nomea eeeceeebxu Pee_eeeeeeeea _eceeeo _ee_eeeeoeeea Peaceeu mmstcm ppm gwnnmm emeeeeem eee new“ ee mommcmmxxoocoz Fesomogowz Fecmm co pcmsummcp pmuwngenocmsa mo pumwwm m mqm

w=cm cm mcw>wmomc azocm mcwpem mzu soce ucmcmmewu apucmowewcmwma .mpmswcm macs Lo Lace mo mm a memos wee mm=Fe> .cowpomncw mewmeopecamo mcwzoFFoe a; we coppwx mam: mpmsece FP< .mcwcwcoFecamo co cowpecpmwcwsue mpmcwm e um>vmumc :owpomncw ume_ Levee a; am new mxeu Lace to» xFFeu moco m:w_mm Lo AmX\ms oov Fmpwncenocmsa mo cowpommcw a? mpmcwm m um>wmumc mpmewcwzcm cm mcw>vmumc azocm Fwo :cou an» ace» ucmgmeewu a—pcmuvewcmwmn .WFmewcm mess to pace eo mm H memos wee mmape> .cowumcpmwcchm mcwuwcop negamo mcvzohpom c; «N umppwx mew: mpmswcm PP< .Fwo :Loo um>wmumg maaogm Focpcoo we» .cowuecpmwcwsee wcwuwgo_m;nmo ou Loved are om cm>wm we: Amx\ms ooopv mewxouan chogma_ae am._nm.__ a~.~nm._P am.m nm.mm aem.onma.e ooom oeexoosm _»eaeoaea K.onm.e e.on_.e m.FFnF.mK m~.onwm._P ooo~ Pee eeeu DP.Fno.e_ aF.Nnm.NN ao.~ am.FN aam.onmw.K coo, oewxoeam Faeocoaaa a.mnm.m N.Nnm.m o.m na.me m~.onep.o_ coop Fee eceu a_._nm.m_ ea.onm._~ m.o ne.em NN.onmN.N o oeexeozm Pxeoeoaea a.mna.em a.mnm.mm m.o nm.e~ mN.onN_.m 0 Pee eeeu z\m _:cm cm mcw>wmomc mum; apes Eocw pcwcmmwwu zpuchwmwcmwm .mpmeNcm mNoE No Lao» we .m. m N we umP—Nx mez cam mcwpmm No mcwcwcopmcamo mo cowpumwcw memos wee mmzpm> .a. N mpmcwm e um>wmumc mPeENc .quep ewe me umNNNx mew: use AmNUN5m>V om cmmzh NF to AmNNmE ova mcvxouaa cmocmawa eo :oNpmNumN:NEcm mpmcwm e vm>wmomc muwnnemm DNN.onNm.N Ne.onNe.e aN.enm.Ne a.mnm.mN aN.ONnm.mNN e.Nnm.eNm ea aNo.onNN.o NN.onNo.N aa..na.moN o.enN.ea no.8 nN.Nm a.mnN.ea_ oe emo.onme.o No.Onee._ aN.Nne.mNN N.enm.Na am.m nN.eN a._nN.mm_ om umammcp pocucou umumwch Pocucou emummcp Pocucou Newsv e=_eoz\ooeNm NNENUaV oeNeNceNeeaoo cacao: NmNmav oeeeeco_aeaou ooeNm oeeN ee_oaa=oeH emmUNNm Neuwucou chmm pwanem :N mxmpa: chnNNo_e;amu co Newspmwcpmga mvwxopsm Facogmnwa co uumemm op m4m .cmwep Lao; mco umppwx mew: mNmswc< .:N_:c_ mx\me com one I New mapon a cosy use mxmu :30» com AFNNc moco apocucouv m:__mm to An? .mNNmE omv Peuwngenocmsa vm>wwumg mpmewc .Nzog mco Low c_e\_e P—.o we zon m be cwpscN NON can I New mz—oa a cosy ace mxev Lao; Noe ANNeu muco ANoeucouv chFem No New .ONNme omv Nmuwncmnocmza um>wmomc mpeepc .Nwwmp meson we umppwx mew: mew mcwwwcopwsaww mo cowuowncw a? wpmcwm m cw>wwuwc mpmemcem aow.o nNm.N_ eNN.onwe.oN am:.0nmo.oN NN.onmm.m ANNNNV o3 NeemNoz NoeeNg OOON oooN oom :Noebeawv o prwsmcmm Amx\msv mcwnNNongawu mmpmm cm zuwwexoh chuNNo_m;nwu mm m4m .Lmuwp mcee; we ewFwa mew: ece mcwewce_e;emo me seepemnce e? wpmcwm e em>wwewg mpwe_cee Nm.eneN.e_ Ne.enNm.eN em.enaN.m_ Nm.enma.mN Ameee e3 NeoNNe: Noee_¥ eeeN oomN eee_ ANOLNeaeV e memEeNee AmemEv werewcepezewu emewz cm xuwewxep wcwewgeFegewu mm m4mewmee we: xpwyweeecee ez .ewpwp meee; em emprex mew: new weepem ee seepewnce mpm:_m e ew>emewe ewe» new meme we eem up mesmee cw eweeeemwe me mpwwe we» ewe mew: mpeeweq .meeueeepm memepeemw; Pecme pee ce weepe xecmweeewe Eewcwpwm ee\e:e m ceseuw> me mpumeem .N— weemee Figure 17 120 .Aooexv m ece I saw: emeweum mew: mceepemm .ew>emmee we: memeeew: Fecemmeuoe use pcmcwEeee mew: mNFmO emeeeeu FmENxeee wee :e :eepmewemmwe ece :eeuepeewe> .Aompv eeeem pemeeeeme Eeeempmm ece u eeEmpe> wee :eee :H .mppme empeeep Peeexeee mew cw :mmm mm: :eepe_eeem> Necewm iwoee .meeeem Newpv ucmweeeme EeeemNmm ece AmmFV peweeeeme m :eEepN> :H .Am+mm+ .emmee we: szNmseeeee ez .ewuwF meee; em ewppwx mew: mew AexNeE oomv meeeeeepeeewe ee :eepOwwcw mecem m em>emewe cmep use mxme we eee eF weeeee CN eweeeemme me mpwee we“ ewe mew: mNeEwe< .mceeeeepeeewe we ex\me com emuee weeueeeum Oeeepeumws Fmeme pee :e zeemwoweme Eeeempwm ee\e:e m :mEmuw> we mpeweem .mF weemwe 121 we oeemee Ie . 4.4““ .u.‘lg. r.\\ §e I- o s... N. a . In his.» 122 (Figures 18B and C). The same dose of cephaloridine oroduced more severe vacuolation accompanied with occasional coagulative necrosis in the group fed the diet deficient in both vitamin E and selenium (Figure 180). A higher dose of cephaloridine (1000 mg/kg) significantly increased kidney weight in both control and deficient groups (Figure 14). It also elevated BUN in the control group; however, the same dose of cephalori- dine markedly increased BUN in the selenium-deficient and both deficient groups (Figure 15). Similarly, 1000 mg/kg of cephaloridine signifi- cantly decreased renal cortical gluconeongenesis; the greatest effect was found in the group deficient in both vitamin E and selenium (Figure 16). In the histopathological evaluation, coagulative necrosis of the proximal tubular cells was a common sign in all four groups; but there were some differences in the distribution and severity of the lesions among the groups. In the control diet group, the high dose of cepha- loridine produced necrotic lesions in the outer cortex; proximal tubular cell vacuolation was frequent but the necrotic process was not prominent (Figure 19A). However, the same dose of cephaloridine also causedsome necrosis in the inner cortex along with swelling and hydropic degener- ated tubular cells in both vitamin E and selenium deficient groups; necrotic areas of the proximal tubules was dispersed in the cortex (Figures 19B and C). Eosinophilic cylinders were occasionally observed in the pars recta of the proximal tubules of the outer medulla in the vitamin E deficient group. The most destructive change was found in the group deficient in both vitamin E and selenium after the high dose of 123 .Neeexv m eee I see: ewcweem mew: mcewuemm .memezwwxo meeewwueee new: mamasw eeweeep we cewmeweee ucweemew sew: ecewe wemmwp ewpeeem: ep ewmcese me: xmeeee we :ewueee weeew .Aeewv eeeee ecwwewwwe Eewcwwwm eee m :wEepw> we» seen :H .xwueee we» cw ewmememwe mew: wwwwo mweeee weswxeee mew we mewee preeeme mueemees .meeeee Aumwv pemwewwwe Eewcmwwm wee Ammwv ucmwowwme m ewseew> :H .pcmwwm we: mmweeee Owueeew: pee xweeee emeee mew cw eceeweee we: wwwme eeweeee weewxeee we cewpeweeee> .Am+mm+ .wmewe ece mxee Ne eew ew meeeww cw eweweemwe me mpwwe wee eww mew: mweswe< .mcweweewezemo we mx\ee coop ewuwe weeueeeum meewepmw; wecme pee :e weewwewwwe Eewcwwwm eeNeee m :wsepw> we muemwwm .mw weeeww o ‘1 \ I .l a . 51‘“ 's _ ~ 7"; ' is 1' | E 4 ~ ’ - . i’i'T I L ,dhb '1. O H ' r 4 g F I | p - . I \ ‘ _‘ . A , - ‘ ‘, ‘7 l _ a v ' l ' a J .1 a 4 . , i ‘. -. -‘ .5 123a "v; ‘ . r-s uv9~ ) ' ~‘ Figure 19 124 cephaloridine. A large portion of cortex was replaced by necrotic tissue and tubular lumens were filled with amorphous cellular debris or serous fluid with occasional deposits of mineral substance (Figure 190). Eosinophilic cylinders were found in Henle's loop, distal tubules and pars recta of the proximal tubules in the outer medulla (Figures 20A and B). 125 Figure 20. Effects of deficiency in both vitamin E and selenium on rat renal histological structure after 1000 mg/kg of cephaloridine. Animals were treated the same as described in Figure 18. The tubular cells in the outer strip of outer medulla showed moderate to severe swelling with variable degree of vacuolation. In addition, necrosis was observed (20A). In the inner strip of outer medulla, proteinous cylinders occluded the lumens of the distal tubules (20B). Sections were stained with H and E (x100). 126 Figure 20 DISCUSSION Drug-metabolizing enzyme inhibitors such as piperonyl butoxide, SKF-525A and cobaltous chloride, as well as inducers such as pheno- barbital, 3-methylcholanthrene and B-naphthoflavone have been widely used to study xenobiotic metabolism and toxicity (Jollow gt_gl,, 1977; Gram, 1980). Using these inducers and inhibitors, investigators ob- tained valuable information regarding whether metabolic activation of xenobiotics is necessary to produce tissue injury. In the present study, pretreatment of rats with piperonyl butoxide signicantly de- creased animal susceptibility to cephaloridine nephrotoxicity (Table 7). A similar result was also observed in the rabbit on the basis of histo- pathological findings (Table 8). The protective effect of piperonyl butoxide on cephaloridine nephrotoxicity has been demonstrated pre- viously in the mouse and rat (McMurtry and Mitchell, 1977). Pretreat- ment of rats and mice with piperonyl butoxide decreased the incidence and severity of the renal proximal tubular necrosis caused by cepha- loridine. Furthermore, McMurtry and Mitchell (1977) reported a pro- tective effect by cobaltous chloride. The results from these inhibitor studies suggested that cephaloridine might have to be metabolically activated in order to cause renal damage. However, this type of result 127 128 can not adequately predict the site where biotransformation takes place. In general, the liver will be the first choice because of its great metabolic capacity. It has been difficult to identify extrahepatic sites of biotransformation jg_gjgg, In order to overcome this problem, phenobarbital has been used as a tool, since this inducer has different effects on hepatic and extrahepatic drug-metabolizing enzyme systems in some species (Uehleke and Greim, 1968; Feuer gt gl,, 1971). In the present study, pretreatment of rabbits with phenobarbital (60 mg/kg daily for four days) significantly increased susceptibility to cephaloridine nephrotoxicity (Table 6). In contrast, a similar treat- ment (80 mg/kg daily for four days) had very little effect on rats (Figure 5). These species-specific effects are consistent with the inducibility of renal mixed-function oxidases by phenobarbital. Evi- dence obtained from the present study (Table 3) and others (Uehleke and Greim, 1968; Feuer gt gl., 1971) have shown that treatment with pheno- barbital increases rabbit renal microsomal cytochrome P450 and cyto- chrome b5 concentrations and mixed-function oxidase (e.g., ethoxycou- marin-O-deethylase, benzphetamine-N-demethylase, arylhydrocarbon hy- droxylase, etc.) activites, but does not alter P450 or b5 concentrations or enzyme activities in the rat kidney. The lack of effect of pheno- barbital on renal toxicity of cephaloridine was also demonstrated in mice (McMurtry and Mitchell, 1977). Interestingly, phenobarbital also showed very little effect on mouse renal mixed-function oxidase activity (Kluwe gt_gl,, 1978). Thus, potentiation of cephaloridine-induced nephrotoxicity seems closely associated with the inducibility of renal mixed-function oxidases by phenobarbital, but not with the inducibility 129 of hepatic mixed-function oxidases since phenobarbital induces hepatic mixed-function oxidases in all three species (Uehleke and Greim, 1968; Feuer gt_gl,, 1971; Kluwe 92.91:, 1978). This correlation indicates that the bioactivation of cephaloridine may take place jg_situ in the kidney. However, phenobarbital treatment has been shown to increase renal blood flow (Ohnhaus and Siegel, 1974) and urinary excretion of xeno- biotics (Ohnhaus, 1972). Treatment with phenobarbital may also alter the pharmacokinetics of cephaloridine and thus vary renal cortical accumulation of the antibiotic, which could lead to a change in the drug toxicity. In the present study, phenobarbital treatment altered pharma- cokinetics of cephaloridine in rabbit kidneys, resulting in significant retention of cephaloridine in rabbit renal cortex (Table 10). In contrast, phenobarbital treatment did not significantly alter pharmaco- kinetics of cephaloridine in rat kidneys (Table 11). Higher renal cortical concentration of cephaloridine in phenobarbital-treated rabbits was not the consequence of a higher serum concentration of cephaloridine (Table 10), but rather was an intrarenal event. It could have been due to a decreased efflux or an increased uptake of cephaloridine or a combination of the two processes. Cephaloridine is known to be actively transported into the proximal tubular cells via an organic anion trans- port system (Tune and Fernholt, 1973). Concomitant administration of an organic anion such as PAH or probenecid with cephaloridine inhibits renal cortical accumulation of cephaloridine and the drug toxicity (Tune gt_gl,, l977a,b). In addition, the organic anion transport system is incompletely developed in newborn rabbits (New gt gl., 1959) and because 130 of this delayed development, the newborn kidneys accumulate less cepha- loridine and the nephrotoxicity of cephaloridine is minor in newborn (Wold gt_gl,, 1977a; Hold and Turnipseed, 1978). Furthermore, multiple administration of penicillin (a substrate for the renal organic anion transport) stimulates the development of organic anion transport (Hirsch and Hook, 1969) which increases the ability of the kidney cells to accumulate cephaloridine and increases nephrotoxicity (Wold and Turnip- seed, 1978). These studies strongly suggest that cephaloridine and PAH are handled by the kidney in a similar manner. However, in the present study, treatment of rabbits with phenobarbital did not increase uptake of PAH in renal cortex either jg vivo or jg_vitro (Tables 15 and 16) even though the same treatment increased renal cortical uptake of cephaloridine (Table 12). Thus, phenobarbital treatment appeared to have different effects on PAH and cephaloridine transport. Unlike PAH, cephaloridine is a zwitterion, having a positive charge on the quarternary nitrogen of the pyridinium substitute on the beta- lactam ring in addition to possessing an anionic carboxyl group. Thus, the renal cation transport system may also influence the movement of cephaloridine into or out of the proximal tubular cells. Hold and Turnipseed (1980) found that pretreatment of rabbits with cyanine (an organic cation inhibitor) delayed the disappearance of cephaloridine from renal cortex jg_yjtg_and this inhibitor also decreased the efflux of cephaloridine from preloaded renal cortical slices jg_gittg, These results suggest that the exit of cephaloridine from cortical cells is dependent on a cation transport system. Since phenobarbital treatment induces mixed-function oxidase activities in rabbit kidneys, like 131 tetrachlorodibenzo-p-dioxin (Fowler gt gl., 1977), the induction may be associated with membrane alteration. This could modify the organic cation transport system and therefore alter the exit of cephaloridine from renal cortical cells. However, phenobarbital treatment had very little effect on renal cortical accumulation of organic cations (e.g., TEA, Table 16). Furthermore, treatment with phenobarbital did not appear to decrease efflux of cephaloridine from renal cortical slices (Figure 6). Piperonyl butoxide is known to inhibit microsomal mixed-function oxidase systems (Anders, 1968; Jaffe gt_gl., 1968; Casida, 1970), but little information is available regarding its effect on renal transport. The data from the present study demonstrated that systemic administra- tion of piperonyl butoxide did not affect renal cortical concentration of PAH (Tables 17 and 18); however, the same treatment markedly reduced renal cortical accumulation of cephaloridine (Table 13). Furthermore, renal cortical slices from piperonyl butoxide treated animals accumu- lated much less cephaloridine compared to those from control animals (Table 14). The cortical slices from the same piperonyl butoxide treatment accumulated almost equal or slightly less PAH than those from the controls (Tables 19 and 20). These jg_tjgg and jg_yjttg results indicated that the effects of piperonyl butoxide on cephaloridine and PAH transport are not the same. In addition, the jg_gjttg_data (Tables 14, 19 and 20) suggested significant differences between cephaloridine and TEA transport. 132 Organic anion (PAH) and cation (TEA) transport are two independent systems existing in the renal proximal tubular cells (Rennick, 1972). These transport processes can be divided into three steps. The first step is transport from the blood into the proximal tubular cells across the peritubular membrane. The second step is intracellular retention and the third step is transport out of the cells across the luminal membrane. The difference between renal cortical accumulation of PAH and cephaloridine has been reported to be in the third step. Tune and his associates (1974; 1975) suggested that cephaloridine, unlike PAH, was not readily transported across the luminal cell membrane and was there- fore retained in renal tubular cells. This concept was further studied and extended by Wold gt_gl, (1979; 1980), who reported that efflux of cephaloridine from cortical tubular cells was markedly inhibited by organic cation inhibitors (e.g., cyanine and mepiperphenidol). 0n the basis of these results and studies from other investigators, they proposed that cephaloridine transport depended on organic anion trans- port for entry into the cortical tubular cells from blood, but, because of the cationic nature of the quaternary nitrogen of the pyridinium side chain, required a cation transport step for exit from cortical tubular cells. They further suggested that organic cation transport appeared to be only moderately efficient for the exit of cephaloridine, which therefore allowed for accumulation and maintenance of high concentra- tions of cephaloridine in the renal cortex. Although a high concentration of cephaloridine can result from inefficient cation transport, alternatively it may be due to 133 intracellular binding. Binding of organic anions to intracellular organelles or cytoplasmic proteins has often been considered as an integral component of the organic anion transport system. Arias and his colleagues (Kirsch 23.11., 1975; Arias gt_gl,, 1976) suggested that the cytoplasmic protein, ligandin (GSH S-transferase B), served as a trans- port carrier of organic anions (e.g., PAH and penicillin) in the kidney. Several lines of evidence supported this hypothesis. The GSH S-trans- ferase fraction isolated from homogenates of renal cortex bound various organic anions including PAH and penicillin (Kirsch gt gl,, 1975). Probenecid administration inhibited the binding of penicillin to renal ligandin following ig_gjtg_injection (Kirsch gt gl,, 1975). Pretreat- ment of animals with enzyme inducers (e.g., TCDD) increased the concen- tration of renal GSH S—transferases as well as urinary excretion, plasma disappearance and renal binding of organic anions (Kirsch gt_gl,, 1975; Arias gt gl., 1976). However, some observations by Pegg and Hook (1977) did not support this hypothesis. Renal GSH S-transferase activities did not parallel the development of PAH transport in neonates. The apparent lack of correlation between PAH transport capacity and GSH S-transferase activity was also observed in several instances including inducer treatment, chronic ammonium chloride acidosis, unilateral nephrectomy, etc. Nevertheless, intracellular binding may be an important compound of net transport of some organic anions. As described previously, the results from the present investigation demonstrated that phenobarbital had different effects on PAH and cepha- loridine transport. Recently, cephalothin (another cephalosporin 134 antibiotic) was reported to bind to GSH S-transferase (ligandin) and to inhibit transferase activities (Ketley gt gl., 1975). Like cephalothin, cephaloridine also inhibited GSH S-transferase activity (Table 5) and this antibiotic may bind to transferases and therefore move slowly out of the kidney cells. Two lines of evidence from the present study support this suggestion. Treatment with phenobarbital increased rabbit but not rat renal GSH S-transferase activity (Table 4). Kirsch gt_gl. (1975) also reported that pretreatment of rats with phenobarbital did not increase renal ligandin concentration but doubled hepatic ligandin concentration. This increase in rabbit renal transferase may be re- sponsible for the higher concentration of cephaloridine accumulated in phenobarbital-treated rabbit kidney. Furthermore, in the untreated animals, rabbit renal cortical GSH S-transferase activity was much higher than rat renal enzyme activity (Table 4). These differences are consistent with the ability of rabbit and rat kidney cells to accumulate cephaloridine. Rabbit renal cortex has been shown to accumulate more cephaloridine than rat renal cortex (Tune, 1975). Thus, these results suggest that GSH S-transferases (ligandins) may be major binding sites for cephaloridine in renal cortical cells. Since both phenobarbital and piperonyl butoxide are modulators of cytochrome P-450 systems and also produce significant alterations in renal cortical accumulation of cephaloridine, it is possible that cyto- chrome P-450 molecules are the intracellular binding sites of cepha- 1oridine. Several lines of evidence support this hypothesis. Pretreat- ment of rabbits with phenobarbital increased kidney cytochrome P-450 135 concentration (Uehleke and Greim, 1968; Table 3) and also enhanced renal cortical accumulation of cephaloridine (Tables 10 and 12). In contrast, phenobarbital pretreatment had no effect on rat kidney cytochrome P-450 concentration (Uehleke and Greim, 1968; Feuer gt gl., 1971) or renal cortical accumulation of cephaloridine (Table 11). In rat kidney, there is no sex-related differences in cytochrome P-450 concentrations (Lit- terst gt_gl,, 1977) and similarly, no sex-related difference in cepha- loridine nephrotoxicity was observed (Table 9). Piperonyl butoxide has been shown to bind to cytochrome P-450 (Matthews 23.21:, 1970; Philpot and Hodgson, 1971; Philpot and Hodgson, 1972) and this interaction may prevent cephaloridine from binding to cytochrome P-450. In the present studies, piperonyl butoxide pretreatment decreased cephaloridine accu- mulation in the renal cortex (Tables 13 and 14). The hypothesis proposed by Wold gt g1. (1980) to explain a high concentration of cephaloridine in renal cortex (described previously) cannot rationalize the effect of phenobarbital and piperonyl butoxide on cephaloridine accumulation, since both phenobarbital and piperonyl butoxide had little effect on PAH or TEA transport. Based on the present results and the information from others, a schematic expression of renal transport of cephaloridine may be suggested (Figure 21). Cephaloridine molecules appear to be actively transported into the proXimal tubular cells from the blood via an organic anion transport system which can be influenced by probenecid, penicillin, etc. After cephaloridine molecules enter the cells the antibiotic may share some binding sites with organic acids. In addition, cephaloridine may bind 136 Figure 21. Schematic diagram of the proposed transport of cephaloridine in renal proximal tubular cells. 6 Pretreatment with penicillin in- creases uptake of cephaloridine from the blood into the tubular cells. Phenobarbital pretreatment increases concentrations of cytochrome P450 and/or GSH transferases, which may bind more cephaloridine. 9=Piperonyl butoxide decreases the binding of cephaloridine to cytochrome P450 and/or GSH transferases. Cyanine decreases efflux of cephaloridine from the tubular cells to the lumen. Probenecid decreases uptake of cephaloridine from the blood into the tubular cells. 137 PHENOBARBITAL %@ ENZYMES (P450 GSH Transierases) 1 flé CEPHA ORIDINE “ CEPH PIPERONYL PROBENECID l BUTOXID Q® PAH POOL 97001 CEPH go/ PENICiILLlN 9 CYANINE BLOOD LUMEN Figure 21 138 to other proteins such as cytochrome P-450 and GSH S-transferase which are probably incapable of reacting with other organic anions. Because of the zwitterion nature, cephaloridine may also react with cation binding proteins (Rennick, 1981). All these possible binding reactions may retain cephaloridine in the cortical tubular cells. Even though the effects of phenobarbital and piperonyl butoxide on cephaloridine-induced nephrotoxicity may, as described previously, result from alterations in renal cortical accumulation of cephaloridine, these effects still can be, at least in part, due to changes in the metabolic activation of cephaloridine. If this process, indeed, occurs in the kidney, the activated metabolites (presumably epoxides) of cephaloridine are very likely to react with intracellular GSH. GSH is present in all types of living cells and is able to conjugate a variety of chemically active compounds including reactive intermediates formed during the metabolism of certain xenobiotics (Mitchell gt_gl,, 1976; Boyland and Chasseaud, 1969). These glutathione conjugates usually are nontoxic and readily excreted. When glutathione content is sufficiently depleted and glutathione conjugation becomes less efficient, some I reactive metabolites will be able to react with vital cellular macro- molecules and cause cell injury and death. Many toxic chemicals such as chloroform (Docks and Krishna, 1976), adriamycin (Olson gt_gl,, 1980), acetaminophen (McMurtry gt gl., 1978; Hassing gt gl,, 1979), bromo- benzene (Jollow gt gl,, 1974) and vinyl chloride (Hefner gt gl,, 1975) have been reported to produce tissue damage associated with depletion of GSH. When intracellular GSH is available, most reactive intermediates 139 are conjugated with GSH. Although reactive intermediates may be de- toxified by other pathways (Bucker gt gl., 1979), glutathione conjuga- tion appears to be the most important one. Tissue damage (e.g., necro- sis) and covalent binding of cellular macromolecules occur only when GSH is depleted. Because of this endogenous protective process, a dose threshold exists for many chemical (e.g., acetaminophen and bromoben- zene) induced tissue injury (Mitchell gt gl., 1973; Jollow gt gl., 1973, 1974). Interestingly, a dose threshold for cephaloridine nephro- toxicity also was observed (Tune gt gl,, 1977a; Hold, 1981). In further support of this view, Mitchell gt_gl, (1977) reported that pretreatment of animals with cysteine decreased cephaloridine renal injury in rats and mice. Since cysteine is a precursor of glutathione, administration of cysteine will increase the availability of intracellular GSH (Leaf and Neuberger, 1947; Boyland and Chasseaud, 1967; Meister and late, 1976; Reed and Orrenius, 1977). Cysteine protection also has been demonstrated in many circumstances of tissue injury produced by poten- tial toxic chemicals such as acetaminophen and bromobenzene (Mitchell gt gl., 1973; Potter gt_gl,, 1974; Jollow gt_gl,, 1974) and these protec- tive effects have been shown to be associated with increased glutathione conjugation. Two possible pathways for GSH conjugation of cephaloridine may exist (Figure 22). First, as suggested by Mitchell gt_gl, (1977), cephaloridine may be activated to an epoxide intermediate and this intermediate can then react with GSH. In addition, the positive charge at the quaternary nitrogen of the pyridinium ring can shift to the para 140 l P450 I Oqu, [QT—<39 GS lGSH lGSH }+(:r¢’:2:::;;:j§>-T-121 F§3-- ::___.> SUE; Figure 22. Possible pathways for the formation of glutathione conjugate(s) of cephaloridine or its metabolite(s). 141 position of carbon atom through n-bond resonance and this new positively charged center, with less steric hindrance, may react with GSH (Figure 22). Both potential GSH conjugates could be less toxic and rapidly excreted. A positive correlation has been shown between GSH depletion and tissue injury produced by many chemicals (Jollow gt_gl., 1974; McMurtry gt gt., 1978). A greater degree of glutathione depletion is associated with a greater magnitude of tissue injury. If, indeed, glutathione conjugation is an important protective pathway against cephaloridine nephrotoxicity, this type of correlation should be ob- served. In the present investigation, the results from 48-hr toxicity studies demonstrated that cephaloridine is more nephrotoxic to rabbits than to rats, and is least toxic to mice (Tables 21-23). These species- different sensitivities to cephaloridine toxicity are consistent with the previous histopathological findings (Atkinson gt_gl,, 1966; Welles gt_gl,, 1966). Taking advantage of these species-related sensitivities, we have investigated changes of tissue GSH in these three species following cephaloridine administration (Figures 7-10, Table 24). Cephaloridine produced a dose-related depletion of GSH in rabbit and rat (Figures 9 and 10) but not mouse renal cortical GSH (Table 24). An apparent reduction in tissue GSH content can be due to a change of tissue water content (e.g., edema). However, the data from the present study (Table 25) indicated small changes (less than 5% of controls) in water content which could not be responsible for the depletion of tissue GSH. This depletion was greatest in rabbits, intermediate in rats, and least in mice, a pattern which is consistent with the species 142 susceptibility to cephaloridine nephrotoxicity. In addition, cephalori- dine had very little effect on GSH concentration of liver (Figure 9) or renal medulla (Figure 10). Previous reports have demonstrated cepha- loridine caused renal cortical injury, especially proximal tubular necrosis, but had little effect on the medulla (Silverblatt gt_gl,, 1970; Atkinson gt gl,, 1966) and the liver (Barza, 1978). All these findings suggested that glutathione depletion in renal cortex is closely associated with cephaloridine nephrotoxicity. This suggestion was further supported by diethyl maleate studies (Tables 27-29). Diethyl maleate is known to react with GSH and will deplete hepatic GSH shortly following administration (Boyland and Chasseaud, 1970). This compound has been used to potentiate xenobiotic toxicity by depleting tissue GSH (Jollow gt_gl,, 1974; Mitchell gt gl., 1973; Harris and Anders, 1980), if tissue toxicity produced by xenobiotics can be prevented by GSH conjugation. In the present study, treatment with diethyl maleate also markedly depleted renal cortical GSH (Table 26) and pretreatment of rabbits and rats with diethyl maleate significantly enhanced cepha- loridine nephrotoxicity. The present results and the cysteine studies by Mitchell gt_gl, (1977) strongly suggests a protective role of gluta- thione against cephaloridine nephrotoxicity. In order to determine further whether an epoxide of thiophene is involved with cephaloridine toxicity, we have used another cephalosporin, cephalothin, which also possesses a thiophene ring but has no pyridinium ring. In contrast to cephaloridine, renal cortical GSH was not reduced following cephalothin administration (Figure 12). Furthermore, using [35$]labelled gluta- thione, in the presence of cephaloridine and required enzymes, we could 143 not detect any glutathione conjugate of cephaloridine. One possibility, although unlikely, is that the analytic methods used still are not sensitive enough to detect the potential metabolites. However, most likely, no glutathione conjugate of cephaloridine can be generated in the kidney. In support of this view, Stewart and Holt (1964) analyzed urine samples from cephaloridine treated patients and did not detect any metabolites. Furthermore, using high pressure liquid chromatography, Hold and Turnipseed (l977b) also did not identify any metabolites. Alternatively, depletion of GSH in renal cortex by cephaloridine can be due to a decrease in glutathione synthesis. The synthesis of glutathione involves two sequential enzymatic steps in which ATP is required (Meister, 1976). As described previously (Introduction sec- tion), Tune gt gt. (1979) reported impaired respiration in renal corti- cal mitochondria isolated from rabbits that had received a toxic dose (200 mg/kg) of cephaloridine. Decreased mitochondrial respiration will reduce the synthesis of ATP and lower its intracellular concentration such that GSH synthesis can be limited, resulting in a decrease in renal cortical GSH. This possibility has been investigated by using another nephrotoxic antibiotic, gentamicin (Barza and Miao, 1977). Gentamicin, like cephaloridine, also is accumulated in the cells of the renal cortex (Luft and Kleit, 1974) and high renal cortical concentrations of this aminoglycoside antibiotic have been shown to inhibit renal cortical mitochondrial respiration (Simmons gt_gl,, 1980). However, gentamicin, unlike cephaloridine, had no effect on renal cortical GSH concentration, even when an acutely lethal dose (1000 mg/kg) was used (Table 30). 144 Thus, decreased ATP synthesis resulted from impaired mitochondrial respiration cannot be the cause of renal cortical GSH depletion produced by cephaloridine. Intracellular GSH concentration results from a dynamic balance between GSH degradation and synthesis, and also is determined through a redox status between GSH and GSSG. A decrease in renal cortical GSH after cephaloridine may be due to an increased oxidation of GSH. The results from the present studies in rabbits and rats clearly demon- strated a decreased cortical GSH with a concomitant increased GSSG shortly following cephaloridine administration (Tables 31 and 32). This alteration in GSH/GSSG ratio appeared to be selective, since the same treatments did not have any significant effect on hepatic GSH and GSSG levels (Tables 31 and 32). Furthermore, total renal cortical gluta- thione (GSH and GSSG) concentrations in cephaloridine-treated animals also increased, indicating some gg ggtg synthesized GSH had been oxi- dized to GSSG. Thus, a shift from GSH to GSSG appeared to be a reason- able explanation for GSH depletion after cephaloridine treatment. The next question is how this can happen. An increase in GSSG concentration can be due to an increased oxidation of GSH, a decreased reduction of GSSG or a combination of both. GSH may be oxidized through thiol- disulfide exchange reactions, but mainly is oxidized by activated oxygen, radicals or peroxides (Flohe and Gunzler, 1976). Thus, in- creased formation of these reactive species in the cells will convert more GSH to GSSG. The question now is "can cephaloridine increase the formation of these reactive species?" As described previously 145 (Introduction section), cephaloridine possesses a pyridinium ring, the specific structure also existing in paraquat. Several lines of evidence (Farrington gt_gl,, 1973; Bus gt_gl,, 1974, 1975; Trush gt gl., 1981) indicated that paraquat undergoes a single electron reduction to form the reduced free radical with NADPH serving as a source of electrons for the reduction. Reduced paraquat radical is rapidly reoxidized by molecular oxygen with formation of oxidized paraquat and superoxide radical. Superoxide radicals may then react with iron, which can attack unsaturated lipids of cell membranes to produce lipid hydroperoxides. Lipid hydroperoxides spontaneously decompose to lipid free radicals, initiating the chain reaction process of lipid peroxidation. Since all these components required for the reduction-oxidation reaction and lipid peroxidation have been shown to be present in the kidney (Monserrat gt gl,, 1969; Grinna and Barber, 1973; Yonaha gt_gl,, 1980; Yonaha and Ohbayashi, 1980), the pyridinium ring of cephaloridine may undergo the similar reduction-oxidation cycle to generate superoxide radicals and then initiate lipid peroxidation (Figure 23). If, indeed, this proposed cascade for cephaloridine metabolism occurs in renal proximal tubular cells, several studies can be conducted to evaluate this hypothesis. A variety of methods have been developed to measure lipid peroxidation (Buege and Aust, 1978; Csallany and Ayaz, 1976; Recknagel and Ghoshal, 1966; Riely gt_gl,, 1974; Dillard gt_gl,, 1977), however most of them cannot accurately determine jg_tjgg_lipid peroxidation at a specific tissue. Measurement of conjugated dienes (Recknagel and Ghosal, 1966; ()XJDNZEI) (V CEPHALORIDINE “APP“ REDUCED NADP CEPHALORIDINE L OXIDIZED CEPHALORIDINE - PEROXIDASEH POLYUNSATURATED GSH \ LIPIDS (LH) SELENIUM . MEMBRANE 6556 I DAMAGE . LIPID LIPID EROX'DASE LIPID "zo/ALCOHOLS PEROXIDES "'" RAP'CMS . LOOH ”0") GSSG GSH ( l \ . GSH . CONJUGATED REDUCTASE DIENEs NADPH NADP LH VITAMIN E k G-6-P J DEHYDROGENASE Figure 23. Proposed mechanism for cephaloridine nephrotoxicity involving lipid peroxidation. 147 Sell and Reynolds, 1969) appears to be a relatively suitable method for this purpose. The results from the present study demonstrated that cephaloridine increased conjugated diene concentration in renal cortex (Table 33 and Figure 13). This increase appeared t; be very selective, because the same treatment did not alter hepatic conjugated diene levels (Table 33). Furthermore, increased conjugated dienes occurred shortly after cephaloridine administration (Table 33 and Figure 13), indicating that lipid peroxidation may be a cause of renal cortical cell injury rather than a result of cell damage. This hypothesis was further sup- ported by the results from the dietary study. Removal of vitamin E and selenium from the diet appeared to potentiate cephaloridine nephro- toxicity (Figures 15-20). Vitamin E is known as a free radical scaven- ger (Myers gt gl,, 1977; Goodman and Hochstein, 1977; Thayer, 1977; Urano and Matsuo, 1976) and presumably acts in the membrane by inter- rupting the free radical Chain reactions of lipid peroxidation (Tappel, 1980). Increased vitamin E concentrations in tissues has been shown to prevent or reduce toxicity of several lipid peroxidation-inducing agents such as carbon tetrachloride (Hafemann and Hoekstra, 1977), adriamycin (Myer gt_gl, 1976), ozone (Fletcher and Tappel, 1973), ethanol (Litov gt_ gl,, 1978) and methyl ethyl ketone peroxide (Litov gt_g1,, 1981). Conversely, vitamin E deficiency increased toxicity of oxygen (Mino, 1973) and paraquat (Bus gt_gl,, 1975) which are shown to be caused by lipid peroxidation. 0n the other hand, selenium has been shown to be an essential constituent of GSH peroxidase (Rotruck gt_gl,, 1973; Flohe gt g],, 1973; Oh gt_gl,, 1974), which is the key enzyme to reduce radical- producing lipid hydroperoxides to stable lipid alcohols (Christopherson, 148 1969; Wendel, 1980) or to remove hydrogen peroxide formed in the cells (McCay gt_gl,, 1976; Hoekstra, 1975). Therefore, GSH peroxidase has been considered a detoxifying enzyme for hydrogen peroxide and organic hydroperoxides. Removal of selenium from the diet significantly reduced GSH peroxidase activities in many tissues including kidney (Hoekstra, 1975) and increased the toxicity of several compounds which were shown to act through lipid peroxidation (Bus gt_gl,, 1975; Wandel and Feuer- stein, 1981). Similarly, in the present study, animals fed a diet deficient in vitamin E and selenium appeared to reduce the ability of converting hydroperoxide and/or hydrogen peroxide to nontoxic metabo- lites and therefore became more susceptible to cephaloridine toxicity. Although lipid peroxidation may be an important mechanism respon- sible for cephaloridine-induced nephrotoxicity, the possible damage due to a dramatic change in renal cortical GSH-GSSG ratios should be con- sidered. A significant number of destructive effects have been reported to be due to alterations of intracellular GSH-GSSG status. Kosower gt g1, (1969) and Harris gt g1, (1971) have studied a thiol-oxidizing agent, diamide, and reported that this compound penetrated the red blood cells and nucleated mammalian cells and stoichiometrically and rapidly oxidized intracellular GSH to GSSG. Addition of diamide to the cells depressed protein synthesis and this inhibitory effect was reversed after the cells regenerated GSH (Harris gt gl,, 1971). This observation was further studied by Zehavi-Willner gt gt, (1971), who found that diamide inhibited both translation and initiation steps of protein synthesis. However, initiation was more sensitive to GSH-GSSG altera- tion than translation. Processes involved in translation recovered 149 after a short lag following regeneration of 40-60% of the original GSH; however initiation processes recovered after regeneration of 70-80% of the original GSH (Zehavi-Willner gt gl,, 1971). Later, using a cell- free preparation (rabbit reticulocyte lysate), Kosower gt g1, (1972) was able to show a direct inhibitory effect of GSSG on protein synthesis. Addition of a small amount of GSSG to the lysate preparation caused a conversion of polysomes to monosomes accompanied with a loss of asso— ciated amino acids, and markedly inhibited the initiation of protein synthesis (Kosower gt_gl,, 1972). Thus, protein synthesis appeared to be closely linked to the intracellular GSH-GSSG status and to be regu- lated by GSSG concentrations in the cells. Cephaloridine markedly increased intracellular GSSG in renal cortex (Tables 31 and 32) and this change may inhibit protein synthesis. It has been shown that glutathione in the reduced (GSH) or in the oxidized (GSSG) form can enter the disulfide-sulfhydryl exchange reac- tions with disulfides (Prot-SSoProt) or sulfhydryl (Prot-SH) groups of proteins resulting in the formation of mixed disulfides (GSS-Prot) (Modig, 1969; Harrap £3,912, 1973; Issac and Binkley, 1977; Mannervik and Axelsson, 1975). Through disulfide-sulfhydryl exchange reactions, a variety of enzymes have been shown to be inhibited by GSSG. Purified glycogen synthetase was inactivated by GSSG and this inactive form could be reactived by the addition of GSH. Inactivation of glycogen synthe- tase was the result of the formation of mixed disulfides between GSSG and the sulfhydryl group(s) of glycogen synthetase (Ernest and Kim, 1973). Similarly, pyruvic kinase was shown to be converted by GSSG to a 150 thermolabile and less active form which had a lower affinity for the substrate phosphoenol pyruvate and the oxidized enzyme was reduced by GSH (Van Berkel gt_gl., 1973). In support of this view, patients with deficiency of glutathione reductase were also deficient in pyruvic kinase (Schroter, 1970; Nowicki gt_gl,, 1972). Glutathione reductase is the key enzyme in the reduction of GSSG to GSH and a deficiency of this enzyme will increase intracellular GSSG concentration, which then inhibits pyruvic kinase activity. GSSG also was reported to inhibit human erythrocyte acid phosphatase activity subsequent to formation of mixed disulfide bonds in the enzyme proteins (Bottini and Modiano, 1964). Consistent with this concept, a lower level of acid phosphatase activity was found in erythrocytes from patients with glucose-6-phos- phate dehydrogenase deficiency (Oski gt gl., 1963). Glucose-6-phosphate dehydrogenase is the important enzyme in the formation of NADPH from NADP Coupled with the dehydrogenation of glucose-6-phosphate, and NADPH is required for the reduction of GSSG. Deficiency of g1ucose-6-phos- phate dehydrogenase in erythrocytes will reduce NADPH concentrations and subsequently increase GSSG concentration, which can then inhibit acid phosphatase activity. Alteration of intracellular GSH/GSSG status has been reported to have a variety of effects on kidney functions. Treatment with diamide specifically inhibited several important enzyme activities, such as protein kinase, Na+,K+-dependent ATPase and glucose-6-phosphatase; inhibition of these enzymes is reversible upon addition of GSH (Pillion gt gl,, 1977a; Leibach gt_gl,, 1978). In addition, treatment with 151 diamide inhibited several renal cortical functions such as gluconeo- genesis (Pillion gt gl,, l977b), amino acid transport (Hewitt gt gl., 1974; Chesney gt_gl,, 1979) and sugar transport (Pillion gt gl., 1975, 1976). Inhibition of these transport systems has been related to intra- cellular GSH-GSSG status. Thus, on the basis of this information, it seems reasonable to suggest that cephaloridine may also alter functions of proximal tubular cells through an alteration in GSH-GSSG status. Although increased lipid peroxidation will increase formation of GSSG, high concentrations of intracellular GSSG do not necessarily follow because a great per- centage of GSSG can be reduced by glutathione reductase with the neces- sary reducing equivalent, NADPH. While GSSG is reduced back to GSH, concomitantly NADPH is converted to NADP. The regeneration of NADPH from NADP is mainly catalyzed by glucose-6-phosphate dehydrogenase coupled with oxidation of glucose-6-phosphate. Rose and coworkers (1976) reported that the hexose monophosphate pathway was stimulated in lung tissue, even when animals were exposed to paraquat for only a very short period. Furthermore, addition of paraquat into the incubation medium increased the hexose monophosphate pathway in lung slices taken from nontreated animals (Rose gt gl,, 1976). This increase was related to paraquat concentration in the lung slices. The stimulation of the activity of the hexose monophosphate pathway indicates a requirement for NADPH. Recently, Witschi gthl, (1977) and Smith gt g1, (1979) showed that the ratio of NADPH to NADP was reduced in lungs taken from para- quat-treated rats. Furthermore, paraquat treatment actually decreased 152 absolute NADPH concentration of the lung (Smith gt gl,, 1979). Although paraquat stimulates the hexose monophosphate pathway, it markedly reduced pulmonary net NADPH concentration, suggesting that the rate of NADPH oxidation by cyclic reduction and oxidation of paraquat is greater than the rate of NADPH generation by the hexose monophosphate pathway. Following paraquat treatment, the intracellular NADPH concentration may fall below that required to sustain vital physiological processes. In support of this view, Smith gt g1. (1979) have shown that some important NADPH-dependent processes such as fatty acid synthesis in lung was inhibited following paraquat poisoning. A decreased intracellular NADPH concentration also can affect the GSH/GSSG ratio. Because NADPH is an important reducing equivalent to convert GSSG to GSH, a reaction cata- lyzed by glutathione reductase, a marked decrease in NADPH will attenu- ate the rate of reduction of GSSG. Although in the present study renal cortical NADPH and NADP concentrations were not measured, the high ratio of GSSG to GSH (Tables 31 and 32) could be the result of a reduced NADPH/NADP ratio. Through the cyclic reduction and oxidation reactions, cephaloridine, like paraquat, may oxidize NADPH and eventually deplete renal cortical NADPH concentration. GSSG was accumulated in renal cortex after cephaloridine (Tables 31 and 32), suggesting that NADPH might be markedly depleted. This marked depletion of NADPH very likely will interfere with normal function of proximal tubular cells. A decrease in the concentration of NADPH may not only debilitate vital physiological processes but render the cells more susceptible to lipid peroxidative injury. As described previously, decreased NADPH will deplete GSH concentration. Depletion of tissue GSH has been shown 153 to result in an increase in lipid peroxidation (Wendel gt gl., 1979; Younes and Siegers, 1980, 1981). This increase is possibly due to the lack of glutathione as a scavenger against free radicals and/or hydro- peroxides. Anundi gt_gl, (1979) showed that glutathione depletion alone could evoke cell damage and they related this effect to an increase in lipid peroxidation. But an enhanced lipid peroxidation was seen only after 80% or more of normal GSH level was depleted (Younes and Siegers, 1981). In the present study, cephaloridine treatment did not deplete renal cortical GSH to that magnitude. However, renal cortex contains various types of cells in addition to proximal tubular cells, where cephaloridine is selectively accumulated, therefore the actual GSH depletion in the proximal tubular Cells should be much greater. Thus, lipid peroxidation in proximal tubular cells may take place after GSH was depleted by cephaloridine treatment. CONCLUSION In conclusion, even though cephaloridine is excreted mainly through glomerular filtration, a significant amount of the drug is actively transported into renal proximal tubular cells via an organic anion transport system. In the proximal tubular cells the drug presumably shares common "binding sites" with organic acids and bases. In addi- tion, cephaloridine may bind to other proteins such as cytochrome P450 and glutathione S-transferase. Unlike the entry of the drug into the cells, cephaloridine appears to be transported out of the proximal tubular cells through an organic cation transport system (Figure 21). Because of these unique transport and intracellular binding systems, high concentrations of cephaloridine are developed selectively in renal proximal tubular cells. Through cyclic redox reaction, the high con- centration of cephaloridine will consume large amounts of NADPH and generate abnormally high concentrations of intracellular lipid hydro- peroxides. A marked depletion of NADPH and increased lipid hydroperox- ides will result in a high intracellular GSSG/GSH ratio (Figure 23). High lipid peroxidation rate, high intracellular GSSG/GSH ratio, low intracellular NADPH or a combination of these changes appeared to be responsible for cephaloridine-induced renal proximal tubular injury (Figure 24). 154 155 .ANNeNxeNeesem: emoeecN -cheNeeNecemO eew mEmNcecemE ewmeeeee wee we SeemeNe ONNeEmeom .em weeeNe m0<_2 / \. \ a zo_._.