Toxicity of ifosfamide and its metabolite ... - Springer Link

26 downloads 45 Views 444KB Size Report
SUMMARY. Renal injury is a common side effect of the chemotherapeutic agent ifosfamide. Current evidence suggests that the ifosfamide metabolite ...
In VitroCell.Dev.Biol.--Animal35:314-317,June 1999 © 1999Societyfor In VitroBiology 1071-2690/99 $05.00+ 0.00

T O X I C I T Y OF I F O S F A M I D E A N D ITS M E T A B O L I T E C H L O R O A C E T A L D E H Y D E IN C U L T U R E D R E N A L T U B U L E CELLS JAMES SPRINGATELKENNETH CHAN, HONG LU, SHERRY DAVIES,ANDMARYTAUB Departments of Biochemistry (M. T.) and Pediatrics (J. S., S. D.), School of Medicine and Biomedical Sciences, State University of New York at Buffalo 14214 and Colleges of Pharmacy and Medicine (11. L., K. C.), The Ohio State University, Columbus, Ohio 43210 (Received 9 July 1998; accepted 16 February 1999)

SUMMARY Renal injury is a common side effect of the chemotherapeutic agent ifosfamide. Current evidence suggests that the ifosfamide metabolite chloroacetaldehyde may contribute to this nephrotoxicity. The present study examined the effects of ifosfamide and chloroacetaldehyde on rabbit proximal renal tubule cells in primary culture. The ability of the uroprotectant medication sodium 2-mercaptoethanesulfonate (mesna) to prevent chloroacetaldehyde-induced renal cell injury was also assessed. Chloroacetaldehyde (12.5-i50 gM) produced dose-dependent declines in neutral red dye uptake, ATP levels, glutathione content, and cell growth. Coadministration of mesna prevented chloroacetaldehyde toxicity while pretreatment of cells with the glutathione-depleting agent buthionine sulfoximine enhanced the toxicity of chloroacetaldehyde. Ifosfamide (1000-10 000 ItM) toxicity was detected only at concentrations of 4000 ItM or greater. Analysis of media collected from ifosfamide-treated cell cultures revealed the presence of several ifosfamide metabolites, demonstrating that renal proximal tubule cells are capable of biotransforming this chemotherapeutic agent. This primary renal cell culture system should prove useful in studying the cause and prevention of ifbsfamide nephrotoxicity. Key words: chloroacetaldehyde; ifosfmnide; sodium 2-inercaptoethanesulfonate; nephrotoxicity; glutathione. itant release of the coproduct chloroacetaldehyde (CAA). We have previously shown that CAA causes renal injury in both the isolated and in viw) perfused rat kidney (20,25). The mechanism by which CAA causes kidney damage is not known. In rat liver artd kidney, CAA toxicity is associated with intracellular glutathione and ATP depletion (18,20). ltosfamide metabolites, including CAA, deplete tissue glutathione stores in humans suggesting a similar method of toxicity (7). In this study, we examined the effect of ifosfamide, CAA, and mesna on kidney cell function, ATP levels, and glutathione content using primary cultures of rabbit renal proximal tubule cells. In addition, the ability of these cells to metabolize ifosfamide was evaluated.

INTRI)DUCTION Ifosfamide is a chemotherapeutic agent that is being used with increasing frequency to treat pediatric malignancies. Initially, ifusfamide therapy was limited by the side eft~ct of severe hemorrhagic cystitis. This complication has been circumvented by the concurrent use of sodium 2-mercaptoethanesulfonate (mesna), a synthetic thiol that combines with reactive ifosfamide metabolites to form stable nontoxic thioether compounds (13,16). However, despite mesna, approximately 40% of ifosfamide-treated children develop a permanent subclinical renal tubulopathy and 5% have a persistent De ToniDebre-Fanconi syndrome (6,16). This syndrome is caused by a generalized dysfunction of renal proximal tubule cells and is defined clinically by excessive urinary excretion of glucose, amino acids, phosphate, bicarbonate, and other solutes handled by this nephron segment. Growth failure, rickets, and progressive renal failure are sequelae of this disorder. tfosfamide is a prodrug that must first be biotranstormed by the cytochrome P450 system before it can exert its therapeutic or toxic effects (16,23,24). Ring hydroxylation produces 4-hydroxyifosfamide (HOIF) which is then converted into the active alkylating agent isophosphoramide mustard (IPM) and acrolein, the putative cause of hemorrhagic cystitis. Ifosfamide also undergoes considerable chloroethyl side chain oxidation yielding N2-dechloroethylifosfamide (N2D) and N:~-dechloroethylifosfamide (N3D) together with eoncom-

MATERIAI£ AND METHODS Isolation and culture procedures. Proximal tubule cells were isolated and cultured as previously described (22). Male 2.0-2.5-kg New Zealand white rabbits (Becken's Farms, Sanborn, NY) were sacrificed by CO2 narcosis. The kidneys with the renal artery intact were immediately removed and washed with sterile culture medium (a 50:50 mixture of Dulbeeeo's modified Eagle's medium and Ham's F12 further supplemented with 15 mM HEPES buffer, 20 mM NaHCQ, 5 gg bovine insulin per ml, 5 gg human transferrin per ml, and 5 X 10 8 M hydrocortisone. The renal artery was cannulated with a sterile, blunted 20-gauge needle and kidneys were peffused with phosphatebuffered saline (PBS) until clear of blood. The kidneys were then peffused with a 0.5% solution of iron oxide until black. The iron oxide is trapped in glomeruli facilitating the subsequent isolation of proximal tubules. After perfusion, kidneys were decapsulated. The cortex was removed, minced into 1cm pieces, and homogenized in a sterile Dounee homogenizer. The resulting homogenate was poured onto a 253-~m mesh screen that was placed in series over a 85-gm screen and washed with culture medium. Tubules and glomeruli retained on top of the 85 gm screen were removed and resuspended in a 50-

'To whom correspondence should be addressed. Present address: Division of Nephrology, The Children's Hospital, 210 Bryant Street, Buffalo, New York 14222. 314

315

IFOSFAMIDE NEPHROTOXICITY ml tube containing cuhure medium and a magnetic stir bar. Glomeruli associated with iron oxide adhered to the stir bar which was then removed. To disrupt basement membrane, the tubules were then incubated with 0.05 mg collagenase per ml and 0.05 mg trypsin inhibitor per ml for 2 min at 23 ° C. After digestion with collagenase, the tubules were washed twice by centrifugation, resuspended in culture medium, and inoculated into 35-ram plastic tissue culture dishes. Culture dishes were maintained at 37 ° C in an incubation chamber with a humidified 5% CO2/95% air environment. Medium was changed after 24 h and every 48 h thereafter. All experiments were conducted 6 - 8 d after plating when confluent monolayers had been achieved. Experimental design. Confluent monolayers of proximal tubule cells were exposed to ifosfamide (1000-10 000 pM) alone or to chloroacetaldehyde (12.5-150 ~M) and mesna (100-1000 ~tM) alone or in combination for 1672 h. Subsequently, neutral red dye uptake, cell glutathione content, cell ATP levels, and cell growth were measured as described below. To determine the influence of cell glutathione content on CAA toxicity, monolayers were pretreated for 24 h with 50 taM of the glutathione synthesis inhibitor buthionine sulfoximine (BSO), exposed to 50 ~tM CAA for 16 h and assayed for neutral red dye uptake. Finally, monolayers were exposed to 1000 taM ifosfamide for 6 h and media collected for metabolite analysis as described below. The concentration of ifosfamide used was approximately 10- to 100-fold greater than that found in patients' serum during ifosfamide treatment (23,25). The concentrations of CAA used (12.5-150 ~tM) compare with serum CAA concentrations ranging from 10-109 ~tM and urine concentrations of up to 220 IxM measured in patients receiving ifosfamide (23,25). Neutral red assay. Cell viability was determined by neutral red dye uptake (4). This dye is taken up by viable cells and stored in lysosomes. Dye is then extracted and uptake quantitated by spectroscopy. A 0.33% (wt/vol) stock solution of neutral red was prepared in PBS, filtered by gravity, and added to culture media to achieve a final concentration of 50 p_g/ml. This medium was then applied to culture plates and incubated for 3 h. Medium was then removed by aspiration and monolayers were gently washed with PBS. Neutral red dye taken up by cells was then extracted by the addition of 1% acetic acid/50% ethanol to each culture plate. After 60 min, the extracted neutral red dye was removed and absorbance at 540 nm was measured. The percentage of neutral red uptake by treated cells was calculated by dividing the As4o in treated cells by the A~0 in control cells and multiplying by 100%. ATP levels. ATP levels were measured with the luciferin-luciferase assay as previously described (2). Cells were solubilized with 0.5 ml 0.5% Triton X-100, acidified with 0.1 ml of 0.6 M perchloric acid, and placed on ice. The cell lysate was diluted with 10 mM potassium phosphate buffer containing 4 mM MgSQ; 0.5 ml of this buffer was added to 1 ml of 50 mM sodium arsenate buffer containing 20 mM MgSO4 to which 25 ~tl of 40 mg luciferin-luciferase per ml was added. Light emission was recorded after 20 s in a beta scintillation counter accepting signals out of coincidence. Cell protein content was determined by the method of Bradford (5) with bovine serum albumin as standard on a portion of the cell sample and ATP levels normalized for protein content of the monolayer. Glucose uptake. Glucose transport studies were performed as previously described (15). Transport medium contained (in mM): 137 NaC1, 4.7 KC1, 0.44 KH2PQ, 1.2 MgSQ, 2.5 CaCI2, 4 glutamine, 10 HEPES buffer, and 0.1 mg bovine serum albumin per ml. To achieve sodium-free medium, sodium was replaced isosmotically by choline. Cell monolayers were washed three times with substrate-free transport medium before incubation in uptake medium containing alpha-[C14]methyl-D-glucoside (0.5 p_Ci/ml), a nonmetabolizable analogue of glucose that shares the apical sodium-dependent glucose pathway in the mammalian proximal tubule. Reactions were terminated at 60 min by our removing solutions and rapidly washing three times with ice-cold isotonic mannitol. Cells were then solubilized in 0.1 N NaOH for 90 min and aliquots sampled for protein and liquid scintillation counting. We calculated sodium-dependent uptakes by subtracting uptake values measured in the presence of choline from those obtained in the presence of sodium. Glucose transport was normalized to cell protein content and expressed as pmol/mg cell protein per unit time. Glutathione content. Total glutathione content was measured in deproteinized cell lysates with a commercial kit (GSH-400, R & D Systems, Minneapolis, MN) and normalized with respect to the protein content of the monolayer. Cell growth determination. Primary rabbit kidney proximal tubule cell cultures treated with either ifosfamide (1000 ~tM) or CAA (25-150 ttM) in either the presence or absence of mesna (1000 ~M) were subcultured by trypsinization as previously described (22). Cells were dislodged by incubation with

A.

Q4 O Q Q

Z

w I

I

I

I

I

I

I

lOO

"

75

50

Z

25

I

I

I

I

I

I

I

0

25

50

75

100

125

150

Chloroacetaldehyde

(uM)

FIG. 1. Effect of increasing concentrations of chloroacetaldehyde on cell growth (A) and neutral red dye uptake (B). For cell growth experiments, treated ceils were inoculated onto 35-mm plastic tissue culture dishes at 10 000 cells/plate. One wk later, ceils were harvested from each plate and counted. Results are means + SE of three experiments. *P indicates a significant difference from the control at P < 0.05. **P indicates a significant difference from results at preceding lower chloroacetaldehyde concentration at P < 0.05.

a 0.05% trypsin-0.5 mM EDTA solution in PBS. Trypsin activity was terminated by the addition of a 0.1% soybean trypsin inhibitor solution in PBS. Harvested cells were washed by centrifugation, resuspended in culture medium, and inoculated onto 35-ram plastic tissue culture dishes at 10 000 cells/plate. One wk later, ceils were harvested from each plate and counted with a Coulter counter model Zf. lfosfamide metabolite analysis. Concentrations of ifosfamide and its metabolites, HOIF, IPM, N2D, and N3D were determined in culture media using a previously described capillary gas chromatography/chemical ionization mass spectrometry procedure (24). Detection limits were 0.04 ~tM ifosfamide, 0.18 pM HOIF, 0.03 ~M N2D, 0.03 taM N3D, and 0.05 IxM IpM. Statistical analysis. Each proximal tubule isolation procedure and culture preparation represented a separate experiment. All studies were performed on at least three separate culture dishes and averaged for each experiment. Results were expressed as mean + SEM of at least four experiments. Means were compared by analysis of variance with the Scheffe post hoc test and considered significantly different when P values were less than 0.05. RESULTS

CAA effects. Treatment with C A A impaired neutral red dye u p t a k e by a n d growth of proximal tubule cells in a d o s e - d e p e n d e n t fashion (Fig. 1). Exposure to i n c r e a s i n g concentrations of C A A p r o d u c e d a progressive i m p a i r m e n t in glucose u p t a k e (Table 1) a n d d e c l i n e s in cell ATP levels (Table 1) a n d glutathione content (Fig. 2). Pretreatm e n t of monolayers with BSO significantly r e d u c e d cell glutathione

SPRINGATE ET AL.

316

TABLE 1 EFFECT OF MESNA ON CHLOROACETALDEHYDE (CAA)-INDUCED PROXIMAL TUBULE CELL TOXICITY° Group

n

Normal CAA (50 pM) Mesna (1,000 ~M) CAA (100 ~tM) CAA (100 p.M) + Mesna (1,000 p.M) CAA (100 laM) + Mesna (500 ktM) CAA (100 ~aM) + Mesna (100 ~tM)

6 6 4 6 4 4 4

Neutral red uptake (%)

74 96 3 97 78 47

100 _+ 3 e + 2 + 2~-" + 2 + 2~ -4- 3~,~

ATP content (nmol/mg protein)

19.8 + 2.5 12.3~ +_ 2.2" 18.0 +_ 2.1 2.5 + 0.9 ~',~ 16.2 + 2.0 no data 5.3 + 1.4~*

Glucose uptake (% of control)

100 79 + 4 ~ 103 + 5 27 + 3 ~'~ 105 + 3 no data 39 + 4 b'~

~Results are means -4-SE. bp < 0.05 vs. normah "P < 0.05 vs. CAA 50 ~tM.

At higher concentrations, cells detached from the culture plates preventing further studies. Analysis of culture media obtained from monolayers treated with 1000 ~tM ifosfamide for 6 h revealed 440 + 22 ~tM ifosfamide, 0.73 + 0.01 ~M OHIF, 2.0 + 0.3 ~tM N2D, 6.2 + 1.3 ~tM N3D, and 1.8 + 0.1 jaM IPM indicating uptake and subsequent biotransformation through both ring and side chain oxidation pathways by proximal tubule cells. Analysis of ifosfamidetreated media not applied to cell cultures revealed no evidence of metabolites.

18 c: =

15

g o o .o x= (.9

6

DISCUSSION

3

CO

C25

C50

C100

M

M+CIO0

FI~;. 2. Ell'eelof chloroacetahlehyde and mesna on renal proximal tubule cell glutathione content. C() indicates control experiments (n - 5) during which no c]dnroacetaldehyde was used. C25 (n = 4), C50 (n = 4), and C100 (n = 4) indicate experiments during which cultures were treated with 25, 50, or 100 Wtl chloruacetaldehyde. M + CIO0 indicates experiments (n = 4) during which cultures were treated with 1000 btM mesna and 100 btM chloroaeetaldehyde. Results are means _+ SE. *P indicates a significant differenee from CO at P < 0.05. **P indicates a significant difference from M at P < 0.05,

levels (control level was 15.3 + 2.8 vs. 8.1 + 1.2/ag protein per mt after treatment with 50 ~M BSO; P < 0.05) and enhanced the toxicity of CAA assessed by neutral red dye uptake (74 + 3% uptake at 50 ~tM CAA vs. 57 + 3% uptake at 50 ~tM CAA plus 50 ~tM BSO; P < 0.05). Mesna effects. Exposure of primary cultures of rabbit proximal tubule cells to 1000 ~tM mesna for 16 h did not significantly affect neutral red dye uptake, glucose uptake, cell ATP levels, or glutathione content (Table 1, Fig. 2). Concurrent treatment with mesna protected monolayers from the toxic effect of CAA in a dose-dependent fashion (Table 1, Fig. 2). Approximately 10-fold more mesna than CAA was required to prevent significant cytotoxicity from CAA. IJbs)hmide-treated cultures. Exposure of primary cultures of rabbit proximal tubule cells to 1000 ~tM ifosfamide for 16 h did not significantly affect neutral red dye uptake, glucose transport, cell growth, ATP levels, or glutathione content (data not shown). Significant reductions in neutral red dye uptake were first detected when monolayers were incubated for 72 h with 4000 ~M itbsfamide (72 _+ 3%).

The present study demonstrates the renal toxicity of the ifosfamide metabolite CAA in vitro and supports previous in vivo evidence of nephrotoxicity (20,25). Chloroacetaldehyde has been previously shown to impair the function and viability of cultured kidney cells and isolated hepatocytes. Exposure of LLC-PK~ cells, a porcine-derived renal epithelial cell line, to more than 50 taM CAA significantly reduces monolayer protein content and inhibits several sodium-dependent transport systems as well as the sodium/proton antiport system {8-10). The mechanism of injury in this renal cell model has not been defined. In hepatocytes, cell damage from CAA is related to glutathione depletion and mitochondrial toxicity with ATP depletion (18,19). Chloroacetaldehyde-induced kidney cell injury in the current study was also accompanied by ATP and glutathione depletion, suggesting a similar pathogenesis. In vivo, this functional damage would cause the de Toni-Debre-Fanconi syndrome, a common manifestation of ifosfamide nephrotoxieity, while total loss of cell viability would lead to acute tubular necrosis and renal failure, an unusual but devastating ifosfamide side-effect (16). The protection oifered by mesna against CAA-induced renal damage in this study contrasts with clinical experience showing that this medication does not eliminate ifosfamide nephrotoxicity (16). After intravenous infusion, mesna is rapidly oxidized in the blood stream to an inactive disulfide, dimesna (13). Both mesna and dimesna are cleared frmn the circulation by glomerular filtration. Dimesna is reabsorbed in the proximal nephron, reduced to mesna and secreted back into the urine where it combines with reactive ifosfamide metabolites to form stable nontoxic thioether compounds, thus preventing hemorrhagic cystitis, The fact that mesna prevents CAA nephrotoxicity in vitro but not in vivo suggests that the intracellular concentration of this uroprotectant is not large enough to neutralize toxic metabolites within renal tubule cells (7,10,11,18). Mohrmann et al. (11)

IFOSFAMIDE NEPHROTOXICITY have shown that LLC-PKa cells are unable to convert dimesna to mesna. Whether primary cultures of proximal tubule cells retain this capacity is unknown. It is of interest in this regard that conversion of dimesna to mesna by proximal tubule cells consumes glutathione, a substance that plays a key role in protecting cells against many types of toxic injury (13). In the present study, exposure of renal tubule cells to CAA produced significant reductions in cellular glutathione content. This result is consistent with previous work showing that ifosfamide metabolites deplete glutathione in renal tissue (12,20). In addition, the toxicity of CAA was enhanced in glutathione-depleted renal cells, a feature shared with other nephrotoxins such as mercury, cadmium and cisplatin (3,14,17). The paradoxical possibility therefore exists that mesna-induced renal glutathione depletion contributes to ifosfamide nephrotoxicity. The above results also provide the first demonstration of renal proximal tubule cell ifosfamide metabolism through both ring hydroxylation and side-chain oxidation pathways and supplement our previous documentation of renal ifosfamide metabolism in the isolated perfused kidney (21). The contribution of this biotransformation to ifosfamide nephrotoxicity is not clear. Although cell viability was affected by ifosfamide in the present study, the concentration of ifosfamide used was quite large compared to that in clinical practice. In LLC-PKa cell cultures, ifosfamide at concentrations of 75-400 ~tmol/ L and exposure times of 1-24 h was found to modestly suppress DNA synthesis without affecting cell viability or transport function (8,9). These findings suggest that renal ifosfamide metabolism does not play a major role in kidney damage. However, renal cell lines can experience a substantial loss of cytochrome P450 activity in culture, and the ability of this cell culture system to quantitatively reflect in vivo conditions is therefore uncertain (1). Further study is needed to define the importance of renal ifosfamide metabolism in this medication's nephrotoxicity. ACKNOWLEDGMENTS This work was supported by grants from The Association for Research of Childhood Cancer, The James H. Cummings Foundation and The National Kidney Foundation of Western New York. REFERENCES 1. Aleo, M.; Taub, M.; Olson, J., et al. Primary cuhures of rabbit renal proximal tubule cells: II. Selected phase I and phase II metabolic capacities. Toxicol. In Vitro 4:727-733; 1990. 2. Andreoli, S.; McAteer, J.; Seifert, S., et al. Oxidant-induced alteration in glucose and phosphate transport in LLC-PK1 cells: mechanisms of injury. Am. J. Physiol. 265:F377-F384; 1993. 3. Bohets, H. H.; Van Thielin, M. N.; Van Der Blest, I., et al. Cytotoxicity of mercury compounds in LLC-PKj, MDCK and human proximal tubule cells. Kidney Int. 47:395-403; 1995. 4. Borenfreund, E.; Puerne~, J. Toxicity determined in vitro by morphological alterations and neutral red absorption. Toxicol. Lett. 24:119-124; 1985.

317

5. Bradford, M. M. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72:248-254; 1976. 6. Ho, P.; Zimmerman, K.; Wexler, L., et al. A prospective evaluation of ifosfamide-related nephrotoxicity in children and young adults. Cancer 76:2557-2564; 1995. 7. Meier, T.; Allenbacher, A.; Mueller, E., et al. Ifosfamide induced depletion of glutathione in human peripheral blood lymphocytes and protection by mesna. Anti-Cancer Drugs 5:403-409; 1994. 8. Mohrmann, M.; Pauli, A.; Ritzer, M., et al. Inhibition of sodium-dependent transport systems in LLC-PK1 cells by metabolites of ifosfamide. Renal Physiol. Biochem. 15:289-301; 1992. 9. Mohrmann, M.; Ansorge, S.; Schmich, U., et al. Toxicity of ifosfamide, cyclophosphamide and their metabolites in renal tubular cells in culture. Pediatr. Nephral. 8:157-163; 1994. 10. Mohrmann, M.; Kupper, N.; Schonfeld, B., et al.. Ifosfamide and mesna: effects on the Na/H exchanger in renal epithelial cells in culture. Renal Physiol. Biochem. 18:118-127; 1995. 11. Mohrmann, M.; Ansorge, S.; Schmich, U., et al. Dithio-bis-mercaptoethanesulphonate (dimesna) does not prevent cellular damage by metabolites of ifosfamide and cyclophosphamide in LLC-PKj cells. Pediatr. Nephrol. 8:458-465; 1994. 12. Nissim, I.: Weinberg, J. M. Glycine attentuates Fanconi syndrome induced by maleate or ifosfamide in rats. Kidney Int. 49:684Mi95; 1996. 13. Ormstad, K.; Orrenius, S.; Lastbom, T., et al. Pharmacokinetics and metabolism of sodium 2-mercaptoethanesulfonate in the rat. Cancer Res. 43:333-338; 1983. 14. Prozialeck, W. C.; Lamar, P. C. Effects of glutathione depletion on the cytotoxic actions of cadmium in LLC-PK, cells. Toxicol. Appl. Pharmacol. 134:285-295; 1995. 15. Sakhrani, L. M.; Badie-Dezfooly, B.; Trizna, W., et al. Transport and metabolism of glucose by renal proximal tubular cells in primary culture. Am. J. Physiol. 246:F757-F764; 1984. 16. Skinner, R.; Sharkey, I.; Pearson, A., et al. Ifosfamide, mesna and nephrotoxicity in children. J. Clin. Oncol. 11:173-190; 1993. 17. Somani, S. M.; Ravi, R.; Rybak, L. P. Diethyldithiocarbamate protection against cisplatin nephrotoxicity. Drug Chem. Toxicol. 18:151-170; 1995. 18. Sood, C.; O'Brien, P. J. Molecular mechanisms of chloroacetaldehydeinduced cytotoxicity in isolated rat hepatocytes. Biochem. Pharmacol. 46:1621-1626, 1993. 19. Sood, C.; O'Brien, P. Chloroacetaldehyde-induced hepatocyte cytotoxicity: mechanisms for cytoprotection. Biochem. Pharmacol. 48:10251032; 1994. 20. Springate, J. Ifosfamide metabolite chloroacetaldehyde causes renal dysfunction in vivo. J. Appl. Toxicol. 17:75-79: 1997. 21. Springate, J.; Zamlauski-Tucker, M.; Lu, H.. et al. Renal clearance of ifosfamide. Drug Metab. Dispos. 29:1081-1082; 1997. 22. Taub, M. Primary kidney cells. In: Pollard, J.; Walker, J., ed. Methods in molecular biology. Vol. 5. Animal cell culture. Clifton, NJ: Humana Press; 1990:189-196. 23. Wagner, T. Ifosfamide clinical pharmacokinetics. Clin. Pharmacokinet. 26:439-456; 1994. 24. Wang, J.; Chan, K. Identification of new metabolites of ifosfamide in rat urine using ion cluster technique. J. Mass. Spectrom. 30:675-683; 1995. 25. Zamlauski-Tucker. M.; Morris, M.; Springate, J. Ifosfamide metabolite chloroacetaldehyde causes Fanconi syndrome in the perfused rat kidney. Toxicol. Appl. Pharmacol. 129:170-175; 1994.