Dosimetry considerations in the enhanced sensitivity of male Wistar rats to chronic ethylene glycol-induced nephrotoxicity

Available online at www.sciencedirect.com

Toxicology and Applied Pharmacology 228 (2008) 165 – 178 www.elsevier.com/locate/ytaap

Dosimetry considerations in the enhanced sensitivity of male Wistar rats to chronic ethylene glycol-induced nephrotoxicity ☆ R.A. Corley a,⁎, D.M. Wilson b , G.C. Hard c , K.E. Stebbins b , M.J. Bartels b , J.J. Soelberg a , M.D. Dryzga b , R. Gingell d , K.E. McMartin e , W.M. Snellings b a

e

Battelle Pacific Northwest Division, Richland, WA 99352, USA b The Dow Chemical Company, Midland, MI 48674, USA c Tairua, 2853, New Zealand d Shell Oil Company, Houston, TX 77002, USA Louisiana State University Health Sciences Center, Shreveport, LA 71130, USA

Received 5 October 2007; revised 26 November 2007; accepted 26 November 2007 Available online 4 December 2007

Abstract Male Wistar rats have been shown to be the most sensitive sex, strain and species to ethylene glycol-induced nephrotoxicity in subchronic studies. A chronic toxicity and dosimetry study was therefore conducted in male Wistar rats administered ethylene glycol via the diet at 0, 50, 150, 300, or 400 mg/kg/day for up to twelve months. Subgroups of animals were included for metabolite analysis and renal clearance studies to provide a quantitative basis for extrapolating dose–response relationships from this sensitive animal model in human health risk assessments. Mortality occurred in 5 of 20 rats at 300 mg/kg/day (days 111–221) and 4 of 20 rats at 400 mg/kg/day (days 43–193), with remaining rats at this dose euthanized early (day 203) due to excessive weight loss. Increased water consumption and urine volume with decreased specific gravity occurred at 300 mg/kg/day presumably due to osmotic diuresis. Calculi (calcium oxalate crystals) occurred in the bladder or renal pelvis at ≥ 300 mg/kg/ day. Rats dying early at ≥ 300 mg/kg/day had transitional cell hyperplasia with inflammation and hemorrhage of the bladder wall. Crystal nephropathy (basophilic foci, tubule or pelvic dilatation, birefringent crystals in the pelvic fornix, or transitional cell hyperplasia) affected most rats at 300 mg/kg/day, all at 400 mg/kg/day, but none at ≤ 150 mg/kg/day. No significant differences in kidney oxalate levels, the metabolite responsible for renal toxicity, were observed among control, 50 and 150 mg/kg/day groups. At 300 and 400 mg/kg/day, oxalate levels increased proportionally with the nephrotoxicity score supporting the oxalate crystal-induced nephrotoxicity mode of action. No treatment-related effects on the renal clearance of intravenously infused 3H-inulin, a marker for glomerular filtration, and 14C-oxalic acid were observed in rats surviving 12 months of exposure to ethylene glycol up to 300 mg/kg/day. In studies with naïve male Wistar and F344 rats (a less sensitive strain), a significant difference was observed in oxalate clearances between young rats (i.e. Wistar clearance b F344) but not in age-matched old rats. Regardless, the ratios of oxalate:inulin clearances in these two strains of rats, including those exposed to ethylene glycol, were all b 1, suggesting that a fraction of the filtered oxalate is reabsorbed. Other species, including humans, typically have clearance ratios N 1 and are more effective at clearing oxalic acid by both glomerular filtration and active secretion. Thus, the lower renal clearance and kidney accumulation of oxalates in male Wistar rats enhances their sensitivity, which will be a factor in human risk assessments. The benchmark dose values (BMD05, BMDL05) were 170 mg/kg/day and 150 mg/kg/day for nephropathy, and 170 mg/kg/day and 160 mg/kg/day for birefringent crystals, using incidence times severity data in each case. The NOAEL of 150 mg/kg/day is the same as that reported after 16-week exposure and appears to be a threshold dose below which no renal toxicity occurs, regardless of exposure duration. © 2007 Elsevier Inc. All rights reserved. Keywords: Ethylene glycol; Oxalic acid; Renal clearance; Chronic toxicity; Nephrotoxicity

☆

The study was sponsored by the Ethylene Glycol Panel, American Chemistry Council. Six co-authors (DMW, KES, MJB, MDD, RG, and WMS) are employees of ACC member companies. ⁎ Corresponding author. Biological Monitoring and Modeling, 902 Battelle Blvd., P.O. Box 999, MSIN P7-59, Richland, WA 99352, USA. Fax: +1 509 376 9064. E-mail address: [email protected] (R.A. Corley). 0041-008X/$ - see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.taap.2007.11.024

166

R.A. Corley et al. / Toxicology and Applied Pharmacology 228 (2008) 165–178

Introduction Ethylene glycol is one of the most well-studied high production volume industrial chemicals due to its decades of use and variety of applications. Although considerable variability has been observed in sensitivity across species, strains and sexes, renal toxicity has long been recognized as a potential outcome following repeated exposures to several hundred mg/kg/day. Ethylene glycol also induces developmental toxicity at higher doses (1000 mg/kg/day) and dose rates (gavage) in rats and mice but not in rabbits. Several recent reviews have summarized the current state of knowledge of the modes of action for these two endpoints (e.g. Carney, 1994; Carney et al., 1999; CERHR, 2003; Corley et al., 2005b). What became clear from these analyses is the critical importance of metabolism and pharmacokinetics in potential susceptibilities and the importance of quantitating these relationships in animal-to-human extrapolations. For example, developmental effects observed in animals are associated with a dose-rate dependent buildup of the intermediate metabolite, glycolic acid, due to saturation of enzymes responsible for further metabolism (Carney, 1994, Carney et al., 1999; Corley et al., 2005b). The initial steps, metabolism of ethylene glycol, accumulation and distribution of glycolic acid, and the metabolic and renal clearance of glycolic acid were therefore incorporated into a physiologically based pharmacokinetic model to improve risk assessments for glycolic acid-induced developmental toxicity (Corley et al., 2005a; Corley and McMartin, 2005). Based upon the now extensive pharmacokinetic and mode of action database for developmental toxicity, there is negligible concern that developmental toxicity will occur in humans under normal use conditions (CERHR, 2003; NTP, 2004). Kidney toxicity, however, occurs in all species studied at dose levels lower than those causing developmental toxicity in rats and mice and thus remains the primary organ of concern for current human health risk assessments. Although notable quantitative differences have been observed, the mode of action for kidney toxicity is essentially the same in animals and humans and involves three key events: (1) the metabolism of ethylene glycol to oxalic acid via glycolic acid; (2) precipitation of calcium oxalate crystals in the kidneys; and (3) degeneration of renal tubule epithelium due to physical trauma or localized oxidative stress (Corley et al., 2005b). Chronic studies have previously been conducted in F344 rats and B6C3F1 mice with rats found to be more sensitive to nephrotoxicity than mice and males more sensitive than females (DePass et al., 1986; NTP, 1993). No treatment-related tumors were observed in either species or sex. For rats, all high dose (1000 mg/kg/day) males died by 15months due to nephrotoxicity with the next lowest dose (200 mg/kg/day) representing a no-observed adverse effect level (NOAEL). Such a dramatic dose–response relationship precludes effective benchmark dose analysis favored by several regulatory agencies.

Risk assessments for chronic exposures have been further complicated by an apparent strain difference in the sensitivity of male rats to ethylene glycol-induced nephrotoxicity suggesting that Wistar rats may be more susceptible. In an unpublished 16-week dietary study, a NOAEL of 71 mg/kg/day, nearly three-fold lower than the chronic NOAEL in F344 rats, was reported in male Wistar rats (Gaunt et al., 1974). However, comparisons of NOAEL's from these two studies are problematic given that the doses used in the chronic study were adjusted periodically to maintain targeted mg/kg/day doses based upon group mean body weights and feed consumption while the subchronic study utilized a constant concentration of ethylene glycol in the diets. Thus, in the Gaunt et al. study, significantly higher dose levels were achieved during the first week of the study than at week 16. As a result, risk assessors are faced with the dilemma of using either the chronic NOAEL of 200 mg/kg/day from the male F344 rat or from the subchronic NOAEL of 71 mg/kg/day from the Wistar rat with an additional uncertainty factor (possibly 10-fold) for lack of a chronic study in this strain, potentially producing a 30-fold difference in reference dose. We have therefore undertaken a series of studies aimed at refining human health risk assessments for lifetime exposures based upon nephrotoxicity as an endpoint and developed a framework for making quantitative comparisons across strains and species. In our subchronic toxicity study, Cruzan et al. (2004) verified that the male Wistar rat is more sensitive than male F344 rats following identical exposures and criteria for pathological evaluation. While the no-observed adverse effect level (NOAEL) for renal toxicity in both male Wistar and F344 rats was 150 mg/kg/day following 16 weeks of exposure when dietary concentrations were adjusted weekly to achieve targeted dose levels, the severity of nephrotoxicity was significantly greater in Wistar rats at 500 and 1000 mg/kg/day, leading to lower benchmark doses in the Wistar rat when incidence and severity data are taken together. Cruzan et al. also included the analysis of blood, urine and kidneys for ethylene glycol and its metabolites, glycolic acid and oxalic acid, following one and sixteen weeks of exposure and demonstrated that the greater nephrotoxicity seen in male Wistar rats vs. the F344 rat was associated with greater kidney oxalate levels, consistent with the mode of action. The purpose of this current study was, therefore, to extend the work of Cruzan et al. (2004) and establish the NOAEL and benchmark dose levels for nephrotoxicity following 12 months of chronic exposure in the most sensitive animal model (male Wistar rat). To assist in dose, route and cross-species extrapolations for human health risk assessments, the concentrations of ethylene glycol, glycolic acid and total oxalates were determined in blood, urine and kidney. In addition, the ability of male Wistar rat kidneys to excrete oxalic acid in the urine was evaluated as a function of age and prior ethylene glycol exposure. Oxalate clearances were also evaluated in agematched naïve male F344 rats. Quantitative relationships were thus developed between male Wistar and F344 rats, and published data in other species including humans, to facilitate comparisons.

R.A. Corley et al. / Toxicology and Applied Pharmacology 228 (2008) 165–178

Methods Study design. Groups of 20 male Wistar Han rats were fed diets formulated to deliver 0, 50, 150, 300, or 400mg ethylene glycol/kg body weight/day for up to 12 months, starting at approximately 6 weeks of age, to evaluate chronic renal toxicity. Ten animals per group were designated for toxicity assessments while five animals per group were pre-selected for analysis of ethylene glycol, glycolic acid and oxalic acid in blood, kidneys and urine after 12 months of dietary exposure to ethylene glycol. The remaining five animals per group were used for the determination of the rates of 14C-oxalate and 3H-inulin clearances, also after 12 months of exposure. All high dose (400 mg/kg/day) animals in the main and satellite groups were terminated early (after 203 days on study) due to excessive mortality and significant depressions in body weights, thus, these animals were not included in the oxalate and inulin clearance studies although kidneys from these animals were saved for metabolite analysis. Chemicals. Polyester grade ethylene glycol used in the toxicity study was supplied by The Dow Chemical Company, Midland, MI (CAS #107-21-1, characterization purity = 99.4% ± 0.07% by gas chromatography with thermal conductivity detection). Ethylene glycol was stored in amber-colored glass bottles at 20 °C purged with nitrogen. Ethylene glycol (Lot No. JR00244CR) and glycolic acid (Lot No. 16802LR) used for analytical standards were obtained from the Aldrich Chemical Company (Milwaukee, WI). Oxalic acid (Lot No. 123H1122), also used for analytical standards was obtained from Sigma (St. Louis, MO). Deuterated internal standards D2-glycolic acid (Lot No. I1-5086) and D4-ethylene glycol (Lot No. P-6136) were obtained from Cambridge Isotope Laboratories, Inc. (Andover, MA) while the internal standard, 2-butoxyethanol (Lot No. 07847HN) was obtained from the Aldrich Chemical Company. Derivatizing reagents, pentafluorobenzoyl chloride (PFBCl) and N-( tert-butyldimethylsilyl)-N-methyltrifluoroacetamide (MTBSTFA) were also obtained from the Aldrich Chemical Company. 14Coxalate (specific activity of 5 mCi/mmol and radiochemical purity of 99%) and 3 H-inulin (specific activity of 103 mCi/mmol and radiochemical purity of N99%) used in clearance studies were obtained from American Radiolabeled Chemicals, Inc. (St. Louis, MO). All other compounds and solvents were reagent grade or better. Animals. Male Wistar Han (Crl:WI(Gix/BRL/Han)IGSBR and F344 (CDF(F344)/CrlBR) rats were obtained from Charles River Laboratories, Inc. (Raleigh, North Carolina). This supplier and specific breeding facility were the same as that used in a previous subchronic toxicity study (Cruzan et al., 2004). Animals were housed individually in stainless-steel mesh caging suspended above cageboard. Drinking water was available ad libitum via pressure-activated, nipple-type watering system, except when water consumption was measured from water bottles at the end of the study. Room temperature was maintained within a range of 21.5–22.3 °C and relative humidity within a range of 48.6–63.0%. Light timers were set to provide a 12-hour light/12-hour dark photoperiod. Animals were fed lower-protein NTP2000 diet in meal form from Zeigler Brothers, Inc., Gardners, PA, ad libitum. NTP2000 diet does not have an excess in protein content and was used to minimize potential confounding by elevated protein of increased incidence and severity of an age-related spontaneous disease (chronic progressive nephropathy) (Rao et al., 1993). Animals were maintained in accordance with the Guide for the Care and Use of Laboratory Animals (1996). The animal care program is accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International (AAALAC International). Diet preparation. Ethylene glycol doses of 0, 50, 150, 300, and 400 mg/kg/ day were based upon results from our previous 1- and 16-week toxicity studies (Cruzan et al., 2004) that suggested that 150 mg/kg/day may be a NOAEL and dose levels of 500 mg/kg/day or greater may be overtly toxic. Diets were prepared by serially diluting a concentrated test material-feed mixture (premix) with ground feed, passing it through a Comil (Quadro Engineering Incorporated, Waterloo, Ontario) to break any small clumps, then further mixed with NTP2000 diet using a Hobart blender. The test diets were prepared weekly and stored under ambient conditions. The concentrations of ethylene glycol in the feed were adjusted weekly for the first 12weeks of the study and at 2-week intervals thereafter, based upon the most recent body weight and feed consumption data. Stability and homogeneity analyses of ethylene glycol in the diet were

167

determined at various time points during the study. Ethylene glycol in feed was analyzed by a solvent extraction method followed by gas chromatography/mass spectrometry and found to be stable in the diets over the duration they were utilized (data not shown). Toxicity assessment. Animals were observed daily for clinical signs of toxicity, and detailed physical examinations were conducted weekly or biweekly. Rats were weighed weekly during the first twelve weeks of the study and at approximately two-week intervals thereafter until test day 190 when significant body weight losses were being observed in the highest dose rats thus, body weights were again recorded weekly through study termination. Animals in the control, 50, 150, and 300 mg/kg/day groups were acclimated to water bottles for approximately three days, eight hours per day, prior to measuring water consumption after 12 months of exposure to ethylene glycol. Water consumption along with urine production was determined during a 24hour period prior to necropsy for all surviving animals (fasted) using glass metabolism cages. Specific gravity (ATAGO Urine Specific Gravity Refractometer), pH, color, and appearance were evaluated. Semiquantitative analysis of pH, bilirubin, glucose, protein, ketones, occult blood, and urobilinogen was also conducted using the Clinitek 200+ (Multistix® Reagent Strips, Bayer Corporation, Elkhardt, Indiana). Microscopic evaluation of crystal types via microsediment analysis was conducted for individual animals. Pathology. After 12 months of treatment, all survivors were euthanized and a complete necropsy conducted. The liver (excluding the 400 mg/kg/day group) and kidneys were weighed and compared to terminal fasted body weights. Onehalf of each kidney was fixed in formalin for pathology assessments, and the remaining half was weighed and flash-frozen in liquid nitrogen for analysis of metabolites as described below. Hematoxylin/eosin-stained kidney sections were examined by normal light (brightfield) microscopy for pathologic lesions, by polarized light for the presence of oxalate crystals (Khan et al., 1982; Rushton et al., 1981), and by fluorescence microscopy for the presence of lysosomes (Maunsbach, 1966; Hard and Snowden, 1991). The severity of compoundinduced nephropathy was graded on a scale of 1–5, and crystal deposition was measured on a scale of 1–4 (see Tables 3 and 4 for a description of the grades). Any gross observations were recorded at necropsy, and any gross lesions along with a full set of tissues were saved in formalin. Since the mode of action of ethylene glycol is well documented (Corley et al., 2005b) and complete histopathological analyses have been conducted on other studies, the focus of the pathology examinations in the current twelvemonth chronic study was the kidney. Based upon observations of urinary bladder stones in some ethylene glycol-treated rats at 300 and 400 mg/kg/day, the urinary bladder from most animals was also examined histopathologically. Similar necropsy procedures were followed for all animals found dead or moribund, except that body weights, organ weights, and urine samples were not obtained. Early termination. As the study progressed, several animals given 400 mg/ kg/day died and the remaining animals at this dose level generally had excessive body weight loss, therefore, the remaining sixteen animals were euthanized on study day 203. Animals were weighed the day prior to euthanasia, fasted overnight, and terminal fasted body weights were collected prior to necropsy. Kidneys from all animals were processed for histologic evaluation and metabolite analysis as described above. Standard tissues and any gross lesions were preserved in formalin. Benchmark dose analysis. Benchmark dose analysis using incidence and severity data for compound-induced nephropathy and birefringent crystal for the purposes of defining a dose corresponding to an extra risk of 5% (BMD05) and its lower confidence limit (BMDL05) was performed using the multistage model (the USEPA MBDS program, version 1.3.2). Metabolite analysis. With the exception of the highest dose group (400 mg/ kg/day), which was terminated early, all animals designated for metabolite analysis after 12 months of exposure were placed in glass metabolism cages for the separate collection of urine and feces on dry ice for 24 h prior to termination (dietary exposures were continued). Following the collection of urine, each metabolism cage was rinsed with water for analysis of ethylene glycol, glycolic

168

R.A. Corley et al. / Toxicology and Applied Pharmacology 228 (2008) 165–178

acid and oxalic acid with the results included with urine as total amounts excreted by this route. After 24 h, each animal was anesthetized under CO2, weighed, heparinized whole blood collected from the vena cava (flash-frozen over dry ice), and humanely euthanized by decapitation. Half of each kidney (one cut longitudinally, one cut transversely) was collected and flash-frozen on dry ice or liquid nitrogen and, along with the blood, urine and cage wash samples, stored frozen at − 80 °C until analyzed for ethylene glycol and its metabolites. The remaining half of each kidney was fixed in formalin for kidney pathological evaluation. In addition, half of each kidney from all animals on the main toxicity study, including those exposed to 400 mg/kg/day and terminated early (day 203), was also collected as above for analysis of ethylene glycol and metabolites. Samples of heparinized whole blood, kidneys, urine and cage wash were analyzed for ethylene glycol, glycolic acid and oxalic acid (total oxalates including calcium oxalate) by gas chromatography/mass spectrometry (GC/MS) following methods previously developed in our laboratories (Carney et al., 1999; Pottenger et al., 2001; Cruzan et al., 2004). Briefly, 0.1g aliquots of whole blood, urine, cage wash or kidney (homogenized with no diluent) were added to 0.9 ml 1N HCl containing internal standard (deuterated ethylene glycol and glycolic acid), and glycolic acid and oxalic acid were extracted using methyl tbutyl ether (MTBE) containing 0.5% tri-n-octylphosphine oxide (2 × 2 ml). The combined MTBE extracts were evaporated to dryness under high purity nitrogen, reconstituted in 0.9 ml toluene and derivatized with 100 μl N-methylN-(t-butyldimethylsilyl)trifluoroacetamide at 60 °C for 1 h. The derivatized samples were analyzed by GC/MS for glycolic acid and oxalic acid and their deuterated internal standard analogues. For ethylene glycol, the aqueous layer remaining after the MTBE extraction of glycolic acid and oxalic acid was neutralized with 200 μl 5M NaOH followed by extractive alkylation of ethylene glycol by the addition of 20 μl pentafluorobenzoyl chloride and 1 ml of toluene (vortex-mixed at 45 °C for at least 30min). The toluene layer was analyzed for ethylene glycol and its deuterated internal standard by GC/MS. Matrix standard curves (control blood, urine or kidneys spiked with known amounts of each analyte and internal standards and processed in the same manner as the samples) were used for quantitation. GC/MS analyses were performed on a Hewlett Packard 7683 Mass Selective Detector equipped with a Hewlett Packard 6890 Plus gas chromatograph and 7683 autosampler (Hewlett Packard, Avondale, PA). Separations were achieved with a Restec RTX-5MS fused silica capillary column (30m × 0.25 mmid, 0.25 μm film thickness; Restec, Bellefonte, PA). Injections (1 μl) were either splitless (GA, OX) or pulsed splitless (EG) using an unpacked Restec 4 mmid cyclo double gooseneck liner. Representative chromatography conditions for glycolic acid and oxalic acid were as follows: injector temperature was 210 °C, the initial oven temperature was 110 °C for 1min, which was then increased at 15 °C/min to 200 °C, with a final ramp of 25 °C/min to 300 °C; initial head pressure was a constant 25psi with helium as the carrier gas. For ethylene glycol, the injection temperature was 190 °C; the initial oven temperature was 130 °C, which was increased at 20 °C/min to 200 °C, with a final ramp of 50 °C/min to 300 °C. The initial head pressure was pulsed at 35psi for 0.5 min followed by a constant head pressure of 25psi with helium as a carrier gas. The mass selective detector was operated in electron impact/full scan mode. The masses used for quantitation of the pentafluorobenzoyl ester derivatives of ethylene glycol and D4-ethylene glycol were 238 and 241, respectively. The masses used for the quantitation of the t-butyldimethylsilyl derivatives of glycolic acid, D2-glycolic acid and oxalic acid were 247, 249 and 261, respectively. For urine samples containing very high concentrations of ethylene glycol, a direct analysis of urine by GC/FID was also conducted using 2butoxyethanol as an internal standard as described in Cruzan et al. (2004). The resulting sample collection (whole blood and kidneys flash-frozen immediately after collection; urine collected over dry ice), storage (− 80 °C), and analytical methods (acidification of sample prior to extraction) minimize artifacts others have discovered to be associated with sample processing that can lead to erroneously high backgrounds of oxalic acid or poor extraction efficiencies (Blau et al., 1998; Braiotta et al., 1985; Chalmers et al., 1985; Hagen et al., 1993; Harris et al., 2004; Hodgkinson, 1981; Mazzachi et al., 1984). Previous studies in our laboratory have demonstrated that whole blood, urine and tissue samples collected, flash-frozen, and stored using these procedures are stable for over one year when samples are kept frozen at − 80 °C (data not shown). All sample analyses from this study were completed within 90 days.

Oxalate clearance. Other than the highest dose group (400 mg/kg/day), which was terminated early, at least 4 of the 5 animals/dose level designated for evaluation of oxalate clearance survived the chronic study. Two additional rats were randomly selected from a sentinel group of animals that were co-housed with the main study animals, and clearance data from these animals were included with the control group. After 12 months of treatment, each animal (under anesthesia) was infused via a jugular cannula with a mixture of 14Coxalic acid and 3H-inulin (a marker for glomerular filtration rates). Blood and urine samples were collected via an abdominal aorta cannula and cannulated bladder, respectively. The methodology used was based on methods of Hautmann and Osswald (1979) and Sugimoto et al. (1993). Oxalate and inulin clearances were also determined in groups of 5 naïve young male Wistar and F344 rats (9–12 weeks of age) and naïve old male F344 rats (47–56 weeks of age) to obtain reference level information on renal function in rats of different strains and ages. Naïve animals were purchased just prior to utilization in clearance studies and therefore, were not on the lower-protein NTP diet that the main study animals were fed. Animals were deprived of food for approximately 14h prior to the clearance study while having free access to water. After anesthesia, the rats were placed on a heated pad to maintain their body temperature and cannulas were implanted for dosing and collection of blood and urine. After an initial infusion of lactated Ringers solution through the jugular vein at a rate of 0.1 ml/min for 0.5h, a solution of oxalate and inulin (0.5–1.0μCi of 14C-oxalate and 0.25–4.5μCi 3Hinulin/ml isotonic saline) was infused with a priming dose at a rate of 0.2 ml/min for 5 min and then a maintenance dose at 0.06 ml/min for 0.5h. After 0.5h of infusion of oxalate and inulin, blood and urine samples were collected for up to 90 min (urine at 10 min intervals with blood collected at the midpoint of each urine collection interval). In some situations, sampling of the blood and urine was delayed until sufficient urine flow had started. Inulin and oxalate concentrations in plasma and urine were determined from 3 H and 14C-radioactivity, respectively, by liquid scintillation counting using a Packard 2900TR scintillation counter (Downers Grove, IL). The clearances of oxalate (CLOX) and inulin (CLIN) were calculated as the product of the urine to plasma concentration ratio and urinary flow rate as follows: CLOX or

IN ðml=minÞ ¼

Urine OX or IN Concentration T Urine Flow ðml=minÞ Plasma OX or IN Concentration ð1Þ

Statistics. Statistical analyses were conducted using SAS/STAT® Software Version 6 or Prism 4 for Windows. Means and standard deviations were calculated for all continuous data. Body weights, feed consumption, organ weights, urine volume, and urine specific gravity were evaluated by Bartlett's test for equality of variances (alpha = 0.01; Winer, 1971). Based on the outcome of Bartlett's test, exploratory data analyses were performed by a parametric (Steel and Torrie, 1960) or nonparametric analysis of variance (ANOVA; Hollander and Wolfe, 1973). If the ANOVA was significant at alpha = 0.05, it was followed, sequentially, by Dunnett's test (Winer, 1971) or the Wilcoxon Rank-Sum test (Hollander and Wolfe, 1973) with a Bonferroni correction for multiple comparisons to the control (Miller, 1966). Metabolite levels were evaluated by a one-way analysis of variance (with log transformation of the data) followed by Dunnett's test to compare each treatment group to controls. Oxalate clearance rates and ratios of oxalate/inulin clearances, with and without normalization to body weights, were evaluated by ANOVA for the Wistar rats in the control and dosed groups. T-tests were used for the following comparisons: young Fischer 344 rats versus old Fischer 344 rats, young Fischer 344 rats versus young Wistar rats, control Wistar rats versus old Fischer 344 rats, and control Wistar rats versus young Wistar rats.

Results In-life There were no treatment-related clinical signs at 50 or 150 mg/kg/day. One control animal and one animal given 150 mg/kg/day died or were terminated early because of

R.A. Corley et al. / Toxicology and Applied Pharmacology 228 (2008) 165–178

spontaneous lymphoid tumors. There were no deaths among rats given ethylene glycol at 50 mg/kg/day. Four rats given ethylene glycol at 300 mg/kg/day died, on days 111, 207, 213, or 221, with a fifth rat at that dose level declared moribund and humanely euthanized on day 138. Four rats given ethylene glycol at 400 mg/kg/day died spontaneously or were humanely euthanized in a moribund state on days 43, 154, 187, or 193. The mortality at 300 and 400 mg/kg/day was considered treatment-related. On day 203, the sixteen remaining animals given 400 mg/kg/ day were humanely euthanized because of excessive body weight loss (Fig. 1). At this dose level, the body weight differences occurred within the first few months but were not statistically identified until day 141, when body weights averaged 12.7% less than controls. This difference progressively worsened until this dose group was terminated. Body weights for rats given 300 mg/kg/day were typically lower but not statistically lower than controls by mid-study. There were no body weight effects at 50 or 150 mg/kg/day. Rats given 400 mg/kg/day had treatment-related decreases in feed consumption at every time point through termination on day 203; the decreases were typically statistically identified from study day 106 (data not shown). There were no treatmentrelated effects on feed consumption for rats given ≤ 300 mg/kg/ day. There was an increase in the incidence of rats having more acidic urine with increasing dose of ethylene glycol (Table 1). The decreased urinary pH was not considered adverse but due to the presence of metabolic products of ethylene glycol. Animals given 300 mg/kg/day appeared to have increased urine volume and concomitantly decreased urine specific gravity compared to controls, which correlated with the increase in water consumption (151% of control). None of these effects reached statistical significance, however, as was observed at higher doses in our previous subchronic study (Cruzan et al., 2004). The more dilute urine in the 300 mg/kg/day group might explain, in part, the finding that fewer animals in this group had decreased urinary pH than in the 150 mg/kg/day group. There were increased

169

Table 1 Water consumption and urine analysis of male Wistar rats administered diets containing ethylene glycol for 12 months Ethylene glycol dose (mg/kg/day)

Water consumption (g/animal/day) Urine volume (ml) Urine pH a Urine specific gravity (mOsmol/L) Incidence of calcium oxalate crystals in urine

0

50

150

300

10.9 ± 3.8

9.7 ± 3.0

8.9 ± 4.0

16.5 ± 8.7

10.6 ± 3.6 8.8 ± 3.7 7.9 ± 3.2 16.3 ± 12.2 0/8 3/10 5/9 1/5 1.031 ± 0.007 1.034 ± 0.009 1.038 ± 0.013 1.025 ± 0.012 1/8

3/10

9/9

5/5

Data are expressed as means ± s.d. except for urine pH and calcium oxalate crystal incidence data. a Urine pH data tabulated as incidence of animals with pH below 7.0. There were no statistically significant effects of treatment (urine pH and calcium oxalate crystal incidence not analyzed).

incidences of calcium oxalate crystals in the urine of the 50, 150, and 300 mg/kg/day groups. This effect was considered a metabolic consequence of ethylene glycol exposure, the low solubility of oxalic acid and the availability of calcium in urine to form calcium oxalate crystals, as no adverse effects in the kidney or bladder were observed in the 50 or 150 mg/kg/day groups (see below). Spontaneous formation of calcium oxalate crystals in the collected urine of naïve laboratory animals is also not uncommon even when care is taken to minimize artifacts (Albasan et al., 2003). Pathology Mean absolute and relative kidney weights (Table 2) in animals given 300 or 400 mg/kg/day appeared to be higher than the control group at necropsy. For example, rats in the 400 mg/ kg/day group were terminated on day 203, yet still had higher mean kidney weights than controls that were terminated on day 369. These increases were not statistically significant at 300 mg/

Fig. 1. Body weights of male Wistar rats exposed to ethylene glycol for 12 months. The mean values for rats given 400 mg/kg/day were significantly ( p b 0.05) reduced compared to control from day 141 onward. The mean values for rats given ≤300 mg/kg/day were not different from control.

170

R.A. Corley et al. / Toxicology and Applied Pharmacology 228 (2008) 165–178

Table 2 Kidney weights of male Wistar rats administered diets containing ethylene glycol for 12 months EG dose (mg/kg/day)

Terminal body weight (g) (n) Kidney weight (g) Relative kidney weight (g/100g body weight)

0

50

150

300

400 a

483.5 ± 39.4 (8) 2.551 ± 0.140 0.530 ± 0.036

498.7 ± 40.3 (10) 2.692 ± 0.297 0.539 ± 0.034

467.6 ± 45.5 (9) 2.417 ± 0.248 0.517 ± 0.028

466.0 ± 93.1 (5) 2.806 ± 0.515 0.612 ± 0.122

367.7 ± 54.5 (16) 4.021 ± 0.746 1.122 ± 0.278

Data are expressed as means ± s.d. (n). a Animals terminated after 203 days. The data for 400 mg/kg/day were not statistically analyzed because there were no concurrent controls. There were no statistically significant effects of treatment for the other dose groups.

kg/day and were not statistically analyzed at 400 mg/kg/day due to the lack of concurrent controls for the early termination of this dose group. Treatment-related gross observations occurred in animals given 300 or 400 mg/kg/day and were primarily confined to the kidney and urinary bladder, with secondary treatment-related observations occurring in the lung. For the 15 rats examined from the 300 mg/kg/day group, 7 had findings in the kidney and 8 had findings in the urinary bladder. For the 20 rats examined from the 400 mg/kg/day group, 17 had findings in the kidney and 10 had findings in the urinary bladder. The most relevant observation in the 300 mg/kg/day group was the presence of calculi in the bladder (and sometimes the renal pelvis or ureter) in 8 of the total 15 rats examined. This also occurred in 8 of 20 rats at 400 mg/kg/day. Calculus formation in the urinary bladder was usually accompanied by dilatation of the bladder and, for the 5 unscheduled deaths at 300 mg/kg/day, hemorrhage of the bladder wall, usually with ascites or other edematous changes. Three animals given 300 mg/kg/day had calculi in the renal pelvis. Almost all rats at 400 mg/kg/day showed gross signs of kidney and/or urinary bladder involvement, including a roughened kidney surface, renal pelvic dilatation, thickened bladder wall, and calculi in the renal pelvis, ureter, or bladder. Of the four unscheduled deaths occurring before early termination of this group, three were observed to have hemorrhage of the bladder wall. All other gross pathological findings were considered secondary to effects observed in the kidneys and bladder, significant body weight loss (400 mg/kg/ day), or agonal changes prior to death. Examination by brightfield microscopy (Table 3) showed that a compound-induced nephropathy associated with crystal deposition affected the majority of the animals at 300 mg/kg/ day and all of those given 400 mg/kg/day. Nephropathy induced by ethylene glycol exposure was observed as foci, radial tracts, or diffuse areas of basophilic tubules in the cortex, and outer and inner medulla. The cytoplasm of basophilic proximal tubule cells was foamy, finely vacuolated, or rarefied, with an occasional apoptotic cell or mitotic figure, and mild basement membrane thickening. There was minimal to mild mononuclear inflammatory infiltration and fibrosis accompanying the basophilic alteration. Increasing severity of the nephropathy was manifested by coalescence of foci into areas of diffuse change and an association with tubule dilatation, increasing fibrosis, increasing extracellular matrix, minor tubulitis associated with intralumenal neutrophils, dilatation of the renal

pelvis, and some transitional cell hyperplasia of the renal pelvic lining. Proximal tubule mineralization was seen in a few advanced cases, but this was not a constant feature. In many kidneys with compound-induced nephropathy, the outlines of crystals could be observed within tubule lumens, or in the renal pelvis, but these were better visualized and scored for severity under polarized light optics. A few rats at the highest dose had either minimal degeneration of the papilla tip, or some pyelitis, both associated with crystal deposition. None of the renal alterations associated with ethylene glycol exposure was observed in the rats given 50 or 150 mg/kg/day, establishing the latter dose as a NOAEL. Under polarized light, birefringent, polycrystalline particles arranged in rosette, fan-shaped or sheaf-like patterns, or individually as near-rectangular plates, were observed in 8 of 13 rats receiving 300 mg/kg/day, and in 10 of 10 rats receiving 400 mg/kg/day (Table 4). Depending on the severity, crystal deposition occurred in the lumens of tubules from the cortex to the papilla, in outpouchings of the papilla lining, and in the renal pelvis, particularly in the fornices. In severe cases the cortex showed frequent crystal deposition. In less severe cases there was only an occasional crystal in the cortex, but more of a concentration in the papilla. In the least affected cases, small crystals were usually observed only in the fornix of the renal pelvis, or in the adjacent urothelial lining. The polycrystalline Table 3 Group incidence and severity of compound-induced crystal nephropathy based on brightfield microscopy in kidneys of male Wistar rats exposed to ethylene glycol for 12 months Dose group (mg/kg/day)

Number of rats assessed

0 50 150 300 400

14 15 15 13 10

Rats with severity grade a 0

1

2

3

4

5

14 15 15 1 0

0 0 0 5 0

0 0 0 2 0

0 0 0 2 1

0 0 0 3 5

0 0 0 0 4

a Crystal nephropathy was graded on a scale of 0 to 5 as follows: 0, no basophilic foci of the type signifying crystal nephropathy; 1, minimal (one to no more than 4 foci of nephropathy in both kidney sections together); 2, mild, (sparsely scattered foci of nephropathy); 3, moderate (frequent foci or early coalescence of foci into areas of nephropathy, but with at least half of the cortex remaining unaffected); 4, marked (diffuse distribution of nephropathy to involve the majority of the parenchyma); 5, end-stage (nephropathy involving all of the kidney indicative of impending renal failure).

R.A. Corley et al. / Toxicology and Applied Pharmacology 228 (2008) 165–178 Table 4 Group incidence and severity of crystal deposition in kidney based on polarized light examination in kidneys of male Wistar rats exposed to ethylene glycol for 12 months Dose group (mg/kg/day)

Number of rats assessed

0 50 150 300 400

14 15 15 13 10

Rats with severity grade a 0

1

2

3

4

14 15 15 5 0

0 0 0 4 0

0 0 0 1 1

0 0 0 2 4

0 0 0 1 5

a Severity grades were as follows: 0, no crystals present in any location in either kidney; 1, minimal (solitary, small crystals evident, usually only in the fornix of the renal pelvis or in the adjacent urothelial lining); 2, mild (usually no crystals in the cortex but scattered in the papilla and renal pelvis); 3, moderate (occasional crystals in the cortex, but more frequent in the medulla); 4, marked (frequent crystals in all zones of the kidney, including cortex).

rosettes and plates had the typical morphology and multicolored birefringence of oxalate crystals (Khan et al., 1982; Rushton et al., 1981). At 300 mg/kg/day, the degree of severity in rats with crystal deposition was 1 (minimal), and at 400 mg/kg/day, the degree was 4 (marked). No birefringent oxalate-like crystals were observed in rats of the control group, or in groups receiving 50 or 150 mg/kg/day. Calculi, up to 2-mm diameter, were found in the bladder, and sometimes in the renal pelvis, at the two highest doses. Since the cause of early death for 3 animals at 300 mg/kg/day was unlikely to be related to the extent of the compound-associated kidney changes, which were less than end-stage, bladder tissue from some animals in each group was examined. Histological findings in the bladder and ureter correlated well with the observations of calculi. The basic change was simple transitional cell hyperplasia, with acute inflammation and hemorrhage in severe cases. In animals dying before scheduled termination in groups given 300 or 400 mg/kg/day, the acute inflammation and hemorrhage of the bladder wall was a consistent finding in all but one case, and considered to be related to the cause of death. Such severe bladder pathology was often accompanied by a necropsy record of ascites or other edematous change, suggesting that infection via the bladder wall and septicemia may have been the terminal event. Benchmark dose analysis Benchmark dose analysis using incidence and severity data resulted in respective BMD05 and BMDL05 values of 170 mg/ kg/day and 150 mg/kg/day for compound-induced nephropathy and 170 mg/kg/day and 160 mg/kg/day for compound-induced birefringent crystals. Metabolite analysis The levels of glycolic acid and total oxalates in whole blood and kidneys of chronically exposed rats are presented in Table 5. Due to the presence of a contaminant in one of the derivatization reagents (PFBCl) that resulted in a significant peak at the same retention time and molecular ions as authentic standards, analysis

171

of parent ethylene glycol in kidneys, blood, and cage wash samples was not completed. Various lots of PFBCl from multiple vendors were screened at the time this study was conducted with similar results [note: PFBCl had been used successfully to derivatize ethylene glycol in prior studies (Pottenger et al., 2001; Cruzan et al., 2004) and again in more recent, ongoing studies; thus, this contamination was an anomaly with unfortunate timing for this study]. Analysis of ethylene glycol in urine was not affected since this procedure did not involve derivatization with PFBCl. At dose levels up to 150 mg/kg/day, there were no differences in the concentrations of glycolic acid and total oxalates in kidney samples, compared with controls with concentrations generally b 2 μg/g (0.026 mmol/kg) and b 20 μg/g (0.22 mmol/kg), respectively. However, at dose levels of 300 and 400 mg/kg/day, both glycolic acid and total oxalates were significantly increased in a dose-related manner. Concentrations at 400 mg/kg/day reached an average of 14 μg/g (0.18 mmol/kg) and 18,789 μg/g (208 mmol/kg) for glycolic acid and total oxalates, respectively, even though these animals were terminated after only 203 days on study; some animals had considerably higher concentrations. The inter-animal variability in total oxalate, and to a lesser degree glycolic acid, levels in kidneys was considerably greater at 300 mg/kg/day than at other ethylene glycol dose levels reflective of the transition between NOAEL (150 mg/kg/day), LOAEL (300 mg/kg/day) and highly toxic (400 mg/kg/day) dose levels. For example, the kidneys from 5 animals in the 300 mg/kg/day treatment group had b 50 μg/g (0.56 mmol/kg) total oxalate while the remaining 5 animals had total oxalate concentrations ranging from 74 to 59,532 μg/g (0.82 to 661 mmol/kg). At 400 mg/kg/day where the incidence of renal toxicity was high, renal oxalate concentrations ranged from 235 to 73,168 23 μg/g (2.61 to 813 mmol/kg) in 15 rats with all but one exceeding 1000 μg/g (11.1 mmol/kg) and 7 exceeding 10,000 μg/g (111 mmol/kg). Higher oxalate levels at 300 and

Table 5 Concentrations (mean ± s.d.) of glycolic acid and total oxalate in whole blood and kidneys of male Wistar Han rats administered ethylene glycol in the diets for up to 12 months Ethylene glycol dose group

Glycolic acid

(mg/kg/day)

(μg/g)

Total oxalate (μg/g)

Blood Control (n = 5) 50 (n = 5) 150 (n = 5) 300 (n = 5)

2.1 ± 1.4 3.4 ± 0.9 2.7 ± 1.9 6.8 ± 1.8 a

3.9 ± 2.4 3.7 ± 2.8 3.8 ± 0.7 5.1 ± 2.2

Kidney Control (n = 13) 50 (n = 15) 150 (n = 14) 300 (n = 10) 400 (n = 15) b

1.7 ± 0.9 1.8 ± 1.0 1.7 ± 1.0 8.6 ± 14.1 a 14.0 ± 9.5 a

5.3 ± 4.2 16.1 ± 35.0 8.7 ± 7.3 6561 ± 18,644 a 18,789 ± 23,446 a

a Statistically different from controls by one-way ANOVA of log-transformed data at alpha = 0.05 followed by Dunnett's test for comparison of individual treatment groups with controls. b Early termination (day 203).

172

R.A. Corley et al. / Toxicology and Applied Pharmacology 228 (2008) 165–178

Table 6 Total amounts (mean ± s.d., n = 5) of ethylene glycol, glycolic acid and total oxalate eliminated in the urine + cage wash collected 24 h prior to termination of male Wistar Han rats administered ethylene glycol in the diets for up to 12 months Ethylene glycol dose group

Amount of urine collected (24-hour)

Ethylene glycol

Glycolic acid

Total oxalate

(mg/kg/day)

(g)

(μg)

(μg)

(μg)

Control 50 150 300

10.1 ± 2.2 15.6 ± 7.4 15.9 ± 4.7 20.0 ± 17.1

ND a 2576 ± 1375 b 6469 ± 1892 b 13,945 ± 8021 b

52 ± 41 232 ± 112 b 359 ± 106 b 2100 ± 1160 b

3015 ± 1486 1519 ± 989 4211 ± 2964 4274 ± 3111

One control urine had detectable amounts of EG (39.8 μg), while no EG was detected (limit of quantitation of 1.54 μg/g) in all other samples. Statistically different from controls by one-way ANOVA of log-transformed data at alpha = 0.05 followed by Dunnett's test for comparison of individual treatment groups with controls. a

b

400 mg/kg/day dose levels were correlated with greater nephrotoxicity (see below). As with the kidneys, the concentrations of glycolic acid in blood were not significantly different from controls up to 150 mg/kg/day (Table 5). At 300 mg/kg/day, the concentrations in blood were approximately 3.2-fold higher than controls although the concentrations were all b 10 μg/g (b 130 μmol/L) regardless of dose level. The concentrations of total oxalate in blood (Table 5) were also similar across all dose levels, averaging 3.7–5.1μg/g (40–60 μmol/L). These results (lack of dose–response in blood oxalate) at the dose levels used in this chronic study were expected from the low solubility of calcium oxalate at physiological pH's in aqueous media (∼ 4.2–7.4μg/g; Burgess and Drasdo, 1993; Hodgkinson, 1981), the lower dose range used in this study (b 10-fold), and previous pharmacokinetic studies of Pottenger et al. (2001) and Cruzan et al. (2004). Background blood oxalate levels were considerably higher in control animals in this study than we have observed in our other shorter term studies where standard diets were used (generally 0–1μg/g [0–10 μmol/L]; Pottenger et al., 2001). However, we have also seen a comparable increase in background oxalate in male F344 and Wistar rats when on the NTP2000 diet for 16 weeks that was not apparent after only 1week (Cruzan et al.,

2004). This higher background appears to be due to this lowerprotein diet or possibly the age of the rats and not sample handling, storage or analytical methods. The total amounts of ethylene glycol eliminated in a 24-hour urine collection at the end of the chronic study followed a linear, dose-dependent increase (Table 6). However, these results will slightly under-estimate the total amounts of ethylene glycol cleared in urine because of the inability to quantitate ethylene glycol in cage wash samples due to the contaminated derivatization reagent as discussed above. Glycolic acid levels were increased over controls at all dose levels with a significant increase in the total amounts excreted observed at the 300 mg/kg/ day dose level vs. the lower two dose levels (Table 6). Total urinary oxalate levels were similar to controls across all dose levels (Table 6). As with blood and kidney data, these results were consistent with those observed in the subchronic study of Cruzan et al. (2004) when similar doses of ethylene glycol (up to 150 mg/kg/day) are compared. Also similar to the blood, the background oxalate levels in control urine samples (3,015μg/24 h or 33.5 μmol/24 h) were higher than we have observed previously in F344 and Wistar rats (0–251μg/24 h or 0– 2.8 μmol/24 h); again, this is attributed to the diet or possibly the age of the animals. This higher background in total oxalates likely

Table 7 Clearances of 14C-oxalate (CLOX) and 3H-inulin (CLIN) in male Wistar Han rats administered ethylene glycol in the diets for up to 12 months and in naïve young male Wistar rats and naïve young and old male F344 rats Ethylene glycol dose group

Body weight

(mg/kg/day)

CLOX

CLIN

Ratio (CLOX/CLIN)

(kg)

(ml/min)

(ml/min/kg BW)

(ml/min)

Male Wistar rat — chronic study Control 50 150 300

0.468 ± 0.024 0.447 ± 0.026 0.449 ± 0.044 0.439 ± 0.042

1.82 ± 0.43 2.00 ± 0.36 2.16 ± 0.44 2.08 ± 0.63

3.91 ± 1.03 4.50 ± 0.59 4.70 ± 0.77 4.79 ± 1.53

2.39 ± 0.64 2.31 ± 0.26 2.99 ± 0.77 3.13 ± 1.78

0.82 ± 0.31 0.87 ± 0.19 0.73 ± 0.08 0.78 ± 0.27

Male Wistar rat — young Control

0.296 ± 0.034

1.13 ± 0.29

3.80 ± 0.70a

2.03 ± 0.70

0.59 ± 0.19

Male F344 rat — young Control

0.193 ± 0.007

1.17 ± 0.13

6.06 ± 0.68 a

1.66 ± 0.16

0.70 ± 0.04

Male F344 rat — old Control

0.395 ± 0.026

1.81 ± 0.54

4.56 ± 1.26

2.15 ± 0.43

0.81 ± 0.10

Values are mean ± s.d. of 4–6 rats/group. a Oxalate clearances statistically different in naïve young male Wistar rats vs. young naïve male F344 rats ( p b 0.001) when normalized to body weight (ml/min/ kg). No other statistically identified differences were observed.

R.A. Corley et al. / Toxicology and Applied Pharmacology 228 (2008) 165–178

obscured dose-depended increases in calcium oxalate which was observed via microscopic analysis of crystals in the urine. Oxalate clearance The average clearances of oxalic acid in subgroups of 4–6 rats/ dose surviving the chronic study ranged from 1.82 to 2.16ml/min with no statistically significant differences observed (Table 7). In addition, no significant differences in treated animals were observed in inulin clearances, which ranged from an average of 2.31 to 3.13 ml/min, based upon ethylene glycol exposure. The only statistically significant difference occurred when the oxalate clearance data were normalized to body weights for naïve young male Wistar (3.80 ± 0.70 ml/min/kg) vs. young male F344 rats (6.06 ± 0.68 ml/min/kg). Although the body weight corrected oxalate clearance remained lower in old male Wistar rats (3.91 ± 1.03 ml/min/kg; control animals from the chronic study) vs. old F344 rats (4.56 ± 1.26 ml/min/kg), the differences were no longer statistically significant. The ratio between oxalate and inulin clearances were all b 1 suggesting that a fraction of filtered oxalate was reabsorbed prior to elimination in the urine. Discussion This study has defined the chronic exposure-internal dose– response relationship for ethylene glycol-induced toxicity in the male Wistar rat. A compound-induced nephropathy associated with crystal deposition affected the majority of the animals at 300 mg/kg/day, and all of those given the highest dose of 400 mg/kg/day with a severity that led to early termination of this group (day 203). In contrast, none of the renal alterations associated with ethylene glycol exposure (basophilic foci of crystal deposition-related nephropathy, tubule dilatation, birefringent crystals particularly in the pelvic fornix as a minimal finding, renal pelvic dilatation, or transitional cell hyperplasia) was observed in the group of rats administered 50 or 150 mg/kg/ day, establishing the latter dose level as a NOAEL. In this regard, the 12-month study recapitulated the results from the 16week study conducted in both Wistar and Fischer 344 male rats, where the dose of 150 mg/kg/day was also an unequivocal NOAEL for both strains (Cruzan et al., 2004). Comparison of these two studies also confirms that there is no progressive or cumulative effect of ethylene glycol with increased duration of exposure at dose levels that were non-toxic in short-term studies as was observed at higher dose levels causing toxicity. In this study, we also demonstrated that the severity of renal effects was associated with the concentrations of total oxalates in the kidney and that renal clearance of oxalic acid is a contributing factor to species differences in sensitivity. The mode of action by which ethylene glycol causes renal toxicity is qualitatively consistent across species and is associated with the metabolism of ethylene glycol to oxalic acid followed by precipitation of oxalic acid with calcium (Corley et al., 2005b). Oxalic acid is poorly soluble in aqueous systems; thus blood levels of oxalic acid in rats typically do not rise above 10–20mg/L (0.1–0.2mM), regardless of the ethylene glycol dose (Burgess and Drasdo, 1993; Hodgkinson, 1981;

173

Pottenger et al., 2001). Thus the concentrations of oxalic acid in blood are not reflective of the dose to the kidneys. This finding was observed in both the subchronic (Cruzan et al., 2004) and current chronic studies. Oxalic acid does not appear to bind to plasma proteins (Corley et al., 2005a) and is freely filtered by the glomerulus. As oxalate ions are concentrated prior to clearance in urine, they can precipitate with calcium ions leading to the growth of insoluble calcium oxalate crystals in the renal tubule epithelium, renal pelvis or bladder. The adherence of calcium oxalate crystals to the epithelium of the kidney tubule can prevent normal kidney function at that site and can lead to cell death. We originally hypothesized that the kidney could become so compromised that it has a reduced capacity to excrete oxalates, resulting in greater kidney burden of oxalic acid and calcium oxalate, further exacerbating the toxicity. As discussed below, we were unable to verify this expectation due to the early termination of the high dose group although we did observe that the male Wistar rat has a reduced capacity for clearing oxalic acid compared with other species regardless of age or duration of ethylene glycol exposure in those animals that survived the chronic study. In the current study, lower pH and calcium oxalate crystals were seen in the urine of Wistar rats dosed at 50, 150, and 300 mg/kg/day, but the critical finding of crystals in target tissues (kidney and bladder) was observed only at dose levels N300 mg/kg/day. The decreased urinary pH is considered a nonadverse result of the formation and clearance of acid metabolites. Doses leading to crystal deposition in the kidney correlated highly with those leading to tissue damage (nephropathy), both occurring at N 300 mg/kg/day. These data are consistent with those reported in the 16-week study of Cruzan et al. (2004) wherein calcium oxalate crystals were seen in the urine of Wistar rats at 150, 500, and 1000 mg/kg/day, but crystals and associated pathology were seen in the kidney only at 500 and 1000 mg/kg/day. For F344 rats on that study, calcium oxalate crystals were also seen in the urine at 500 and 1000 mg/ kg/day, but were only seen in the kidney parenchyma of 1 of 10 rats at 500 mg/kg/day, and in all 10 at 1000 mg/kg/day. Thus, calcium oxalate crystals can be present in the urine without being detected in the renal tubule lumen, and without kidney pathology. Furthermore, spontaneous formation of calcium oxalate crystals in the collected urine of naïve laboratory animals is not uncommon (Albasan et al., 2003). One difference between the subchronic (16-week) and the present chronic (12-month) studies was the finding at necropsy of calculi, up to 2mm in diameter, in the bladder, and sometimes in the renal pelvis, at the two highest doses in the 12-month study. Bladder tissue from animals in this study was examined because the cause of death of the three animals dying at 300 mg/ kg/day was likely unrelated to the extent of compoundassociated kidney changes, which were less than end-stage. Histological findings in the bladder and/or ureter correlated well with the necropsy observations of calculi. Although the cause of death may have been related to the consequences of calculi in the bladder, the most sensitive markers of the adverse effects of ethylene glycol were in the kidney, namely crystal nephropathy and/or the presence of birefringent crystals.

174

R.A. Corley et al. / Toxicology and Applied Pharmacology 228 (2008) 165–178

Calculus formation as a consequence of ethylene glycol administration is a predictable finding given the chronic duration of exposure. In their two-year bioassay of ethylene glycol in Fischer 344 rats, DePass et al. (1986) reported the presence of oxalate crystals in the urinary bladder by 12 months, and occasional calculi in the pelvic space, ureters, and bladder by 18months. The greater sensitivity of the Wistar rat as demonstrated by Cruzan et al. (2004) may explain the more rapid development of calculi by 12 months in the present study. In a subchronic study of calcium oxalate crystalluria induced by ethylene glycol in the Sprague-Dawley strain of rats, Khan (1995) also described the formation of “ministones” on the surface of the renal papilla after 8week s, and referred to the potential for this to lead to stone development. On the basis of the crystalline structures observed in some of the bladders in the current twelve-month study with ethylene glycol, it seems likely that the calculi diagnosed at necropsy were not true concretions, which are usually solid, but merely organization of crystal clumps into larger aggregates. In the DePass et al. (1986) 2-year study with F344 rats, exposure to 1000 mg/kg/day resulted in oxalate crystal nephropathy leading to the death of all male rats before the end of the 24-month study. However, no oxalate crystal nephropathy was seen in rats dosed at 200 mg/kg/day for 24months. In the current 12-month study, the benchmark dose values using incidence and severity data for compound-induced crystal nephropathy based on brightfield microscopy of the kidneys resulted in respective BMD05 and BMDL05 values of 170 mg/kg/day and 150 mg/kg/day for compound-induced nephropathy. The BMDL05 value for the current study reflects the application of the multistage model to data covering a small range of doses (50–400 mg/kg/day) resulting in a good fit with tight confidence limits. The current chronic study may be considered suitable for human health risk assessment of ethylene glycol based upon a consideration of: (1) use of a more sensitive test strain and gender (male Wistar); (2) use of a chronic exposure duration (1year); (3) use of a large number of animals/ group (15); and (4) use of four dose groups over a narrow dose range (50–400 mg/kg/day) to provide a more complete characterization of the response range with respect to incidence (0–100%) than was obtained in prior studies. The internal dose–response relationships for ethylene glycol, glycolic acid and total oxalate in blood, urine and kidneys from animals from the chronic study were also consistent with the previous 1- and 16-week studies of Cruzan et al. (2004) where doses were comparable. However, hyperoxalemia and hyperoxaluria were not readily apparent, primarily due to the lower dose levels in this chronic study, the early termination of the highest dose group (400 mg/kg/day), and the higher background of oxalates in blood and urine that was attributed primarily to the NTP2000 diet. Regardless, the disposition of total oxalate levels in the kidney, the target tissue, clearly shifted between non-toxic (b 150 mg/kg/day) and nephrotoxic (N150 mg/kg/day) dose levels in both the chronic and subchronic studies. Consistent with the mode of action, kidneys from animals with the higher total oxalate concentrations were the most severely affected. Since oxalic acid is an endogenous metabolite and its metabolic precursors or oxalic

acid itself is present in several dietary constituents, background levels of total oxalate are expected in normal, healthy kidneys. In this study, the average total oxalate level was 10.0± 21.5μg/g (0.11 ± 0.24 mmol/kg) with an upper 95% confidence level of 16.8μg/g (0.19 mmol/kg) in all rats from the control, 50 and 150 mg/kg/day dose groups. However, at the nephrotoxic dose levels of 300 or 400 mg/kg/day ethylene glycol, the oxalate levels increased exponentially with dose. In fact, total oxalate concentrations, when expressed as calcium oxalate, accounted for an average of 2.7% of the total kidney weight (with one animal approaching 10.6%) in the animals exposed to 400 mg/kg/day and terminated early (day 203) in the study. For those kidneys which were analyzed for total oxalate as well as scored histologically for treatment-related nephrotoxicity, a log–linear relationship was observed between total oxalate and nephrotoxicity scores (Fig. 2). Furthermore, total oxalate levels in all rats with at least a minimal treatment-related nephrotoxicity score were at or above the 95% upper confidence level for unaffected kidneys. Since urinary clearance is an important factor in kidney oxalate levels, the potential for chronic exposures to ethylene glycol to affect the clearance of oxalic acid in urine was investigated in this study. However, no statistically significant differences were observed between treated animals and controls in either oxalate or inulin clearances, even at a minimalmoderate nephrotoxic dose level of 300 mg/kg/day (Table 7). Unfortunately, the highest dose level, 400 mg/kg/day, was terminated early due to increased morbidity and mortality, so potential effects on renal function, which were anticipated, could not be determined at this dose level where toxicity was more severe. Interestingly, the renal clearance of oxalic acid also appears to be unaffected in humans with underlying kidney diseases such as urolithiasis, glomerulonephritis, or polycystic kidney disease (Boer et al., 1985). However, this may also be due to the tremendous reserve capacity of the kidney to function even when significant percentages of nephrons are damaged. While ethylene glycol, at the dose levels used in this chronic study (up to 300 mg/kg/day), did not affect the urinary clearance of oxalic acid, there was a notable strain difference detected when clearances were normalized to body weight in young, naïve, male Wistar vs. F344 rats (Table 7). Such a difference between strains of young rats likely contributed to the greater dose-related increases in kidney oxalate levels observed in male Wistar vs. F344 rats in shorter term studies with ethylene glycol (Cruzan et al., 2004). These strain differences may have been maintained for a significant period of time although after 12 months, the differences were no longer statistically significant. Lack of strain differences in oxalic acid clearance in naïve old rats suggests that differences previously observed in renal toxicity (Gaunt et al., 1974; DePass et al., 1986; Cruzan et al., 2004) are related to early events associated with oxalate crystal formation and deposition followed by nephrotoxicity. Consistent with previous data in male Wistar rats (Costello et al., 1992), the ratios of oxalate to inulin clearances were b 1 for all male Wistar and F344 rats in this study (Table 8). This ratio indicates that a fraction of the filtered oxalate was reabsorbed by the kidneys in both strains of rat. This is the opposite effect that has been shown to occur in other species, including humans

R.A. Corley et al. / Toxicology and Applied Pharmacology 228 (2008) 165–178

175

Fig. 2. Dose–response relationship between total oxalate and compound-related nephrotoxicity scores in kidneys where both endpoints were evaluated. Individual animals in the 0, 50 and 150 mg/kg/day groups with a nephrotoxicity score of 0 are designated with a triangle (Δ); the upper 95% confidence level for total oxalate in those kidneys (16.8 μg/g) is indicated by the dashed line. Treatment-related nephropathy scores of 1 or greater were only observed in animals exposed to 300 mg/kg/day and terminated after 12 months (♦, diamond) or 400 mg/kg/day and terminated on day 203 due to increased morbidity and mortality ( , square). Scoring legend: 0 = No compound-related nephropathy 1 = Minimal (one to no more than 4 basophilic tubule foci or radial tracts in both kidney sections together) 2 = Mild (sparsely scattered basophilic tubule foci or tracts) 3 = Moderate (frequent foci, or coalescence into areas of nephropathy, but with at least half the cortex remaining unaffected) 4 = Marked (diffuse distribution of nephropathy to involve the greater proportion of parenchyma) 5 = End-stage (nephropathy involving all of the parenchyma).

(Table 8). Of those studies that are most similar in design to the current study (e.g. healthy animals or humans that were infused with inulin and 14C-oxalic acid) the ratio of oxalate to inulin clearance was N 1, indicating that in addition to filtration, the kidneys secreted an additional amount of oxalic acid into the

urine. Similar results were reported from studies where the more variable endogenous marker of glomerular filtration, creatinine, was used instead of inulin infusion (Table 8). Other published studies that were not included in this comparison were those that either relied upon measurements of endogenous oxalate

Table 8 Summary of in vivo studies describing the clearance of 14C-oxalic acid (CLOX) and its relationship to glomerular filtration rate (GFR, ml/min) in healthy humans and animals Species

N

BW

CLOX

GFR

CLOX/GFR

(kg)

(ml/min)

(ml/min)

Ratio

Studies where 14C-OX injected or infused along with inulin as marker for GFR Human a 12 70 249.2 ± 46.0 103.9 ± 16.0 2.34 ± 0.47 6 40 73.9 ± 7.8 53.6 ± 4.0 1.35 ± 0.19 Sheep b Dog c 6 10 68.0 ± 12.0 54.0 ± 15.0 1.28 ± 0.20 Male SD rat 82 0.305 2.73 ± 0.16 2.23 ± 0.13 1.23 ± 0.04 Male F344 rat Male Wistar rat

9 11

0.305 0.390

1.53 ± 0.36 1.51 ± 0.37

1.93 ± 0.31 2.23 ± 0.67

0.76 ± 0.06 0.72 ± 0.26

References

Hautmann and Osswald (1979); Osswald and Hautmann (1979) McIntosh and Belling (1975) Cattell et al. (1962) Knight et al. (1979a,b); Sugimoto et al. (1993); Weinman et al. (1978); Hautmann and Osswald (1978) Current study, combined young and old control rats from Table 7 Current study, combined young and old control rats from Table 7

Studies where 14C-OX injected or infused with GFR estimated from endogenous creatinine in plasma and urine Human d 27 68.6 237.9 ± 40.6 117.4 ± 22.4 2.05 ± 0.23 Boer et al. (1985); Prenen et al. (1979); Prenen et al. (1982) Dog e 4 18.5 71.3 ± 14.2 NA e NA Duggan et al. (1979) Male Wistar rat 16 0.625 1.45 ± 0.16 1.82 ± 0.13 0.80 f Costello et al. (1992) Data are expressed as the weighted means and standard deviations across studies. a Human subjects were either male (n = 5) or sex not specified (n = 7). b Merino ewes. c Mongrel female dogs. d Males (n = 18) and females (n = 9). e Female beagles; GFR not reported. f Data not provided for calculation of standard deviation.

176

R.A. Corley et al. / Toxicology and Applied Pharmacology 228 (2008) 165–178

(generally using enzyme assays) and creatinine, assuming steady state kinetics and estimates of urine flows, or dietary manipulations of oxalic acid. These studies are not experimentally similar to the current study and, more importantly, clearance values calculated from such studies are significantly more variable and less reliable than 14C-oxalate and 3H-inulin infusion studies. Allometric relationships were therefore derived from the studies summarized in Table 8 between body weight (BW, kg) and clearance (Fig. 3) using the standard exponential equation: Clearance ðml=minÞ ¼ aðBWÞb

ð2Þ

For oxalate clearance, the allometric equation was derived only for those species having a common mechanism of secretion of oxalic acid in addition to filtration (e.g. CLOX/GFR N 1;

Fig. 3a). This resulted in a coefficient, a, of 7.0635ml/min/kg with an exponent of 0.7729. For the male Wistar and F334 rats, an apparently different process, tubule resorption, is involved in renal clearance, therefore data from these strains were included in Fig. 3a as a comparison. For inulin clearance, all species and strains were included in the allometric analysis. In this case, the coefficient, a, was 5.1213ml/min/kg with an exponent of 0.7328, which is nearly identical to other published values for glomerular filtration (Adolph, 1949; Hackbarth et al., 1982; Singer, 2001) which gives us greater confidence in the results for oxalic acid. These allometric relationships can be used to predict clearances of either oxalate or inulin based upon body weight for those species having a common mechanism. These relationships can also be used to estimate quantitative differences between

Fig. 3. Allometric relationship between body weight and renal clearance of (a) injected or infused oxalic acid (14C-OX) and (b) inulin in healthy animals and humans (data from Table 8). The regression analysis included only those animals with CLOX/GFR ratios N1; data for F344 and Wistar rats which have CLOX/GFR ratios b 1 are included for comparison.

R.A. Corley et al. / Toxicology and Applied Pharmacology 228 (2008) 165–178

male Wistar rats and other species for use in risk assessments. For example, as shown in Table 8, a 390g male Wistar rat has an oxalate clearance of 1.51 ± 0.37ml/min and a ratio of oxalate to inulin clearance 0.72 ± 0.26. If, for example, oxalate clearance processes in the Wistar rat were similar to other species, including humans (e.g. CLOX/GFR N 1), it would have been expected to have a clearance of 3.41 ml/min based upon this allometric analysis. Thus, when compared with other species, male Wistar rats are more than 2-fold less efficient at clearing oxalic acid and are therefore at increased risk for accumulating oxalate crystals and nephrotoxicity in chronic studies. Given the importance of oxalate levels in the kidneys as a key event in the nephrotoxicity mode of action of ethylene glycol, such a comparison can contribute to adjustments in default uncertainty factors in human health risk assessments based upon chronic nephrotoxicity in the male Wistar rat. In addition to differences in renal clearance, proximal tubule cells from male Wistar and F344 rats bind and internalize more calcium oxalate than proximal tubule cells from humans (McMartin and Guo, 2007), which could contribute to increased tissue loading of oxalates. Furthermore, proximal tubule cells from these two strains of rats are considerably more sensitive to calcium oxalate-induced cytotoxicity than human proximal tubule cells at comparable concentrations in vitro (Guo and McMartin, 2005). Such in vitro pharmacokinetic and pharmacodynamic relationships along with the in vivo target tissue dose–response and clearance analyses conducted in this study will eventually facilitate the development of predictive models for the species-specific dosimetry and potential toxicity of calcium oxalate. In summary, the current study expanded upon the results of our subchronic study (Cruzan et al., 2004) by refining the exposure-internal dose–response relationships for ethylene glycol-induced kidney toxicity after chronic exposure in male Wistar rats, which are known to be a highly sensitive sex and strain of animal from shorter term studies. The determination that 150 mg/kg/day represents a NOAEL for the chronic toxicity of ethylene glycol given orally via the diet to male Wistar rats was based on the absence of systemic, bladder, or renal toxicity, and equates well to the NOAEL observed after only 16 weeks of exposure previously reported. Identical NOAEL's of 150 mg/kg/day for the subchronic and chronic studies indicate that there is a threshold dose for renal toxicity below which exposure of any duration will not be expected to result in adverse renal effects. Quantitative relationships between kidney oxalate and pathology and the renal clearance of oxalic acid across species will further inform human health risk assessments based upon chronic toxicity in the male Wistar rat. Acknowledgments This study was sponsored by the American Chemistry Council CHEMSTAR Ethylene Glycol/Ethylene Oxide Panel. Members of the Panel include: BASF Corporation, the Dow Chemical Company, Eastman Chemical Company, Equistar Chemicals, L.P., Huntsman Corporation, and Shell Chemical,

177

L.P.; Mr. William Gulledge is the Panel Manager. Thanks to Dr. Chris Kirman, The Sapphire Group, for the benchmark dose analysis. Thanks to Drs. Steven M. Green and Lynn H. Pottenger for the critical review of this manuscript. References Adolph, E.F., 1949. Quantitative relations in the physiological constitutions of mammals. Science 109, 579–585. Albasan, H., Lulich, J.P., Osborne, C.A., Lekcharoensk, C., Ulrich, L.K., Carpenter, K.A., 2003. Effects of storage time and temperature on pH, specific gravity, and crystal formation in urine samples from dogs and cats. JAVMA 222 (2), 176–179. Blau, N., Matasovic, A., Lukasiewicz-Wedlechowicz, Heizmann, C.W., Leumann, E., 1998. Simultaneous determination of oxalate, glycolate, citrate, and sulfate from dried urine filter paper spots in a pediatric population. Clin. Chem. 44, 1554–1556. Boer, P., Prenen, J.A.C., Koomans, H.A., Dorhout Mees, E.J., 1985. Fractional oxalate clearance in subjects with normal and impaired renal function. Nephron 41, 78–81. Braiotta, E.A., Buttery, J.E., Ludvigsen, N., 1985. The effects of pH, temperature and storage on urine oxalate. Clin. Chim. Acta 147, 31–34. Burgess, J., Drasdo, D.N., 1993. Solubilities of calcium salts of dicarboxylic acids in methanol–water mixtures; transfer chemical potentials of dicarboxylate anions. Polyhedron 12, 2905–2911. Carney, E.W., 1994. An integrated perspective on the developmental toxicity of ethylene glycol. Reprod. Toxicol. 8, 99–113. Carney, E.W., Freshour, N.L., Dittenber, D.L., Dryzga, M.D., 1999. Ethylene glycol developmental toxicity: unraveling the roles of glycolic acid and metabolic acidosis. Toxicol. Sci. 50, 117–126. Cattell, W.R., Spencer, A.G., Taylor, G.W., Watts, R.W.E., 1962. The mechanism of the renal excretion of oxalate in the dog. Clin. Sci. 22, 43–51. NTP–CERHR expert panel report on the reproductive and developmental toxicity of ethylene glycol. NTP–CERHR-EG-03(May) http://cerhr.niehs.nih.gov2003. Chalmers, A.H., Cowley, D.M., McWhinney, B.C., 1985. Stability of ascorbate in urine: relevance to analysis for ascorbate and oxalate. Clin. Chem. 31, 1703–1705. Corley, R.A., McMartin, K.E., 2005. Incorporation of therapeutic interventions in physiologically based pharmacokinetic modeling of human clinical case reports of accidental or intentional overdosing with ethylene glycol. Toxicol. Sci. 85, 491–501. Corley, R.A., Bartels, M.J., Weitz, K.K., Soelberg, J.J., Gies, R.A., Thrall, K.D., 2005a. Development of a physiologically based pharmacokinetic model for ethylene glycol and its developmentally toxic metabolite, glycolic acid, in adult rats and humans. Toxicol. Sci. 85, 476–490. Corley, R.A., Meeke, B., Carney, E.W., 2005b. Oxalate crystal-induced renal tubule degeneration and glycolic acid induced dysmorphogenesis: mode of action case study on the renal and developmental effects of ethylene glycol. CRC Crit. Rev. Toxicol. 35, 691–702. Costello, J.F., Smith, M., Stolarski, C., Sadovnic, M.J., 1992. Extrarenal clearance of oxalate increases with progression of renal failure in the rat. J. Am. Soc. Nephrol. 3, 1098–1104. Cruzan, G., Corley, R.A., Hard, G.C., Mertens, J.W.M., McMartin, K.E., Snellings, W.B., Gingell, R., Deyo, J.A., 2004. Subchronic toxicity of ethylene glycol in male Wistar and F344 rats is related to metabolism and clearance of metabolites. Toxicol. Sci. 81, 502–511. DePass, L.R., Garman, R.H., Woodside, M.D., Giddens, W.E., Maronpot, R.R., Weil, C.S., 1986. Chronic toxicity and oncogenicity studies of ethylene glycol in rats and mice. Fundam. Appl. Toxicol. 7, 547–565. Duggan, D.E., Walker, R.W., Noll, R.M., VandenHeuvel, W.J.A., 1979. Determination of urinary oxalic acid and of oxalate pool size by stable isotope dilution. Anal. Biochem. 94, 477–482. Gaunt, I.F., Hardey, J., Gangolli, S.D., Butterworth, K.R., Lloyd, A.G. (1974). Short-term toxicity of monoethylene glycol in the rat. Carshalton, Surrey, UK: BIBRA International; Research Report 4/1974. Guide for the Care and Use of Laboratory Animals, 1996. National Research Council. National Academy Press, Washington, D.C.

178

R.A. Corley et al. / Toxicology and Applied Pharmacology 228 (2008) 165–178

Guo, C., McMartin, K.E., 2005. The cytotoxicity of oxalate, metabolite of ethylene glycol, is due to calcium oxalate monohydrate formation. Toxicol. 208, 347–355. Hackbarth, H., Buttner, D., Gartner, K., 1982. Intraspecies allometry: correlation between kidney weight and glomerular filtration rate vs. body weight. Am. J. Physiol., Regul. Integr. Comp. Physiol. 242, 303–305. Hagen, I., Walker, V.R., Sutton, R.A.L., 1993. Plasma and urinary oxalate and glycolate in healthy subjects. Clin. Chem. 39, 134–138. Hard, G.C., Snowden, R.T., 1991. Hyaline droplet accumulation in rodent kidney proximal tubules; an association with histiocytic sarcoma. Toxicol. Pathol. 19, 88–97. Harris, A.H., Freel, R.W., Hatch, M., 2004. Serum oxalate in human beings and rats as determined with the use of ion chromatography. J. Lab. Clin. Med. 144, 45–52. Hautmann, R., Osswald, H., 1979. Pharmacokinetic studies of oxalate in man. Invest. Urol. 16, 395–398. Hautmann, R., Osswald, H., 1978. Renal handling of oxalate. A micropuncture study in the rat. Naunyn-Schmiedeberg's Arch. Pharmacol. 304, 277–281. Hodgkinson, A., 1981. Sampling errors in the determination of urine calcium and oxalate: solubility of calcium oxalate in HCl–urine mixtures. Clin. Chim. Acta 109, 239–244. Hollander, M., Wolfe, D.A., 1973. Nonparametric Statistical Methods. John Wiley, New York. Khan, S.R., 1995. Calcium oxalate crystal interaction with renal tubular epithelium, mechanism of crystal adhesion and its impact on stone formation. Urol. Res. 23, 71–79. Khan, S.R., Finlayson, B., Hackett, R.L., 1982. Experimental calcium oxylate nephrolithiasis in the rat. Role of the renal papilla. Am. J. Pathol. 107, 59–69. Knight, T.F., Senekjian, H.O., Taylor, K., Steplock, D.A., Weinman, E.J., 1979a. Renal transport of oxalate: effects of diuretics, uric acid, and calcium. Kidney Int. 16, 572–576. Knight, T.F., Senekjian, H.O., Weinman, E.J., 1979b. Effect of paraaminohippurate on renal transport of oxalate. Kidney Int. 15, 38–42. Maunsbach, A.B., 1966. Observation on the segmentation of proximal tubule in rat kidney. Comparison results from phase contrast, fluorescence, and electron microscopy. J. Ultrastruct. Res. 15, 239–258. Mazzachi, B.C., Teubner, J.K., Ryall, 1984. Factors affecting measurement of urinary oxalate. Clin. Chem. 30, 1339–1343. McIntosh, G.H., Belling, G.B., 1975. An isotopic study of oxalate excretion in sheep. Aust. J. Exp. Biol. Med. Sci. 53, 479–487.

McMartin, K.E., Guo, C., 2007. Binding and internalization of oxalate crystals by proximal tubule cells from humans and rats is related to oxalate cytotoxicity. The Toxicologist 96 (S-1), 264–265. Miller, R.G. Jr. (1966). Simultaneous Statistical Inference. McGraw-Hill, New York, pp. 67–70, 101–102. NTP. (2004). NTP–CERHR monograph on the potential human reproductive and developmental effects of ethylene glycol. National Institutes of Health, National Toxicology Program; NIH Publication No. 04-4481, January, 2004. NTP. (1993). Toxicology and carcinogenesis studies of ethylene glycol (CAS No. 107-21-1) in B6C3F1 mice (feed studies). Technical Report 413; National Institutes of Health, National Toxicology Program; 1993. Osswald, H., Hautmann, R., 1979. Renal elimination kinetics and plasma halflife of oxalate in man. Urol. Int. 34, 440–450. Prenen, J.A., Boer, P., Mees, E.J., Endeman, H.J., Spoor, S.M., Oei, H.Y., 1982. Renal clearance of [14C]oxalate: comparison of constant-infusion with single-injection techniques. Clin. Sci. 63, 47–51. Prenen, J.A.C., Dorhout Mees, E.J., Boer, P., Endeman, H.J., Ephraim, K.H., 1979. Oxalic acid concentrations in serum measured by isotopic clearance technique. Experience in hyper- and normo-oxaluric subjects. Proc. Eur. Dial. Transplant Assoc. 16, 566–569. Pottenger, L.H., Carney, E.W., Bartels, M.J., 2001. Dose-dependent nonlinear pharmacokinetics of ethylene glycol metabolites in pregnant (GD10) and nonpregnant Sprague-Dawley rats following oral administration of ethylene glycol. Toxicol. Sci. 62, 10–19. Rao, G.N., Edmondson, J., Elwell, M.R., 1993. Influence of dietary protein concentration on severity of nephropathy in Fischer-344 (F-344/N) rats. Toxicol. Pathol. 21 (4), 353–361. Rushton, H.G., Spector, M., Rodgers, A.L., Hughson, M., Magura, C.E., 1981. Developmental aspects of calcium oxalate tubular deposits and calculi induced in rat kidneys. Invest. Urol. 19, 52–57. Singer, M.A., 2001. Of mice and men and elephants: metabolic rate sets glomerular filtration rate. Am. J. Kidney Dis. 37, 164–178. Steel, R.G.D., Torrie, J.H., 1960. Principles and Procedures of Statistics. McGraw-Hill, New York. Sugimoto, T., Osswald, H., Yamamoto, K., Kanazawa, T., Iimori, H., Funae, Y., Kamikawa, S., Kishimoto, T., 1993. Fate of circulating oxalate in rats. Eur. Urol. 23, 485–489. Weinman, E.J., Frankfurt, S.J., Ince, A., Sansom, S., 1978. Renal tubular transport of organic acids. Studies with oxalate and para-aminohippurate in the rat. J. Clin. Invest. 61, 801–806. Winer, B.J., 1971. Statistical Principles in Experimental Design, (2nd Edition). McGraw-Hill, New York.