Factors Affecting Inorganic Mercury Transport and Toxicity in the Isolated Perfused Proximal Tubule1

Transcription

1 actors Affecting Inorganic Mercury Transport and Toxicity in the Isolated Perfused Proximal Tubule1 Rudolfs K. Zalups,2 Mary K. Robinson, and Delon W. Barfuss R.K. Zalups, Division of Basic Medical Sciences, Mercer University School of Medicine, Macon, GA M.K. Robinson, Division of nvironmental Health Laboratory Sciences, Center for nvironmental Health and Injury Control, Centers for Disease Control, Atlanta, GA D.W. Barfuss, Biology Department, Georgia State University, Atlanta, GA (J. Am. Soc. Nephrol. 1991; 2:86#{243}-878) ABSTRACT The effects of cysteine (8 MM), glutathione (8 MM), rabbit albumin (1MM), and an ultrafiltrate of rabbit plasma on the toxicity and transport of inorganic mercury (Hg24 ; 18.4 MM) in isolated perfused SI, S2, and 53 segments of the renal proximal tubule from the rabbit were studied. Cellular and tubular injuries were assessed qualitatively by light microscopy observations and quantitatively by the tubular leak of the volume marker 3H-glucose. The lumen-to-bath transport of inorganic mercury was assessed by measuring both the rate of disappearance of inorganic mercury from the luminal fluid and the rate of appearance of inorganic mercury in the bath. When glutathione was added to the perfusate containing the Inorganic mercury, no signs of epitheliol cell necrosis or injury were detected in any of the three segments of the proximal tubule. There was also an absence of or a decrease in cellular injury in the epithelium of the same tubular segments when either cysteine or the ultrafiltrate was present in the perfusate. However, when rabbit albumin and inorganic mercury were present in the perfusate, severe degenerative and necrotic changes occurred very rapidly in the epithelium of all three segments of the proximal tubule. In almost every instance where glutathione, cysteine, or the plasma ultrafiltrate were present in the perfusate, the disappearance flux of Inorganic mercury from the tubular lumen into the Received January 14, Accepted May Correspondence to Dr. IlK. Zolups, Division of Basic Medical Sciences, Mercer University School of Medicine, 155 College Street, Macon, GA /24-866$3./ Journal of the American Society of Nephroiogy Copyflghto 1991 by the American Society of Nephrology tubular epithelium was lowered. It was concluded that glutathione, cysteine, and the ultrafiltrate of rabbit plasma provide isolated perfused SI, S2, and S3 segments of the proximal tubule varying degrees of protection from the toxic effects of inorganic mercury. This protection appears to be related to a decrease In the movement of inorganic mercury across the luminal membrane of the tubular epithehal cells. Key Words: Cysteine. glutathione. albumin, plasma ultrafiltrate. kidney, proximal tubule, rabbit, inorganic mercury toxicity and transport I n a recent study, we examined the toxicity and transport of inorganic mercury (Hg2) in isolated perfused 51, S2, and S3 segments of the proximal tubule from the rabbit kidney (1). We discovered that when inorganic mercury is delivered into the lumen of any of the three segments of the proximal tubule (at concentrations between 1 nm and greater than 1 MM) in a simple electrolyte solution that does not contain any compound possessing free sulfhydryl groups, cellular degeneration and necrosis occur very rapidly at the end of the tubular segment attached to the perfusion pipette. Our previous findings also show that when 18.4 MM inorganic mercury is perfused through the lumen of the three segments of the proximal tubule, inorganic mercury is taken up very rapidly by the tubular epithelial cells in these segments. The inorganic mercury that is absorbed and retained by the proximal tubular epithelial cells appears to end up being bound to cellular proteins associated with the epithelial cells. Unfortunately, we do not know whether any free ionic mercury enters the lumen of any of the segments of the proximal tubule in the In vivo situation where the whole animal has been exposed to a dose of inorganic mercury. We do know, on the other hand, that when rabbits are given a single Iv nontoxlc dose of inorganic mercury, nearly 5% of the dose of inorganic mercury is present in the total renal mass within 48 h after the dose is administered (2). Most of the inorganic mercury is present in the 51, S2. and S3 segments of the proximal tubule (2). It is not known, however, whether the inorganic mercury Is taken up from the luminal membrane, basolateral membrane, or both. What is of particular importance is the fact that the urinary excretion of inorganic 866 Volume 2 Number 4 #{149} 1991

2 Zalups ef al mercury is a slow process. which is not well understood. Our previous findings show that only about 6 to 8% of a nontoxic dose of inorganic mercury is excreted in the urine during the Initial 48 h after the dose is administered (2). Soon after a dose of inorganic mercury is administered, very little of the dose is present in the blood (2). Approximately 5% of the fraction of inorganic mercury in the blood is in the red blood cells (3). The remainder is in the plasma, and it is mainly bound to the free sulfhydryl groups of plasma peptides and proteins, primarily albumin (4.5). Recent histochemical findings tend to indicate that some Inorganic mercury In plasma is filtered at the glomerulus and is taken up by the epithelial cells in all three segments of the proximal tubule by pinocytosis (6,7). Data obtained from In vitro filtration experiments indicate that only about 1 to 2% of the mercury in plasma can filter at the site of the glomerulus (unpublished findings by R.K. Zalups). These data were obtained from experiments where the plasma of rats and rabbits treated with 23Hg2 was filtered in a small filtration cell possessing a filtration membrane having a molecular mass cutoff of 5. Da. Assuming that Inorganic mercury is in fact filtered at the glomerulus, then what form is the inorganic mercury in when it Is filtered? Is the mercury bound to compounds containing free sulfhydryl groups, is it In the free ionic form, or is it a combination of both? Because a number of compounds containing free sulfhydryl groups (such as glutathione, cysteine, and albumin) are filtered at the site of the glomerulus, and because inorganic mercury has a great affinity for sulfhydryl groups on peptides and proteins, it would seem highly probably that some of the inorganic mercury in blood does filter at the glomerulus in a form bound to the sulfhydryl groups of some of these compounds. Thus, it is important to evaluate the transport and toxicity of inorganic mercury in the three segments of the proximal tubule under conditions comparable to those in plasma ultrafiltrate as well as In the presence of some individual compounds containing free sulfhydryl groups. In the study presented here, we examine the effects of glutathione. cysteine, albumin, and a plasma ultrafiltrate from the rabbit on the transport and toxicity of inorganic mercury in the isolated perfused Si. S2, and S3 segments of the proximal tubule from the rabbit kidney. MATRIALS AND MTHODS Animals New Zealand White rabbits of varying sizes were used in this study. The animals were housed and maintained according to the National Institutes of Health guidelines for the care and use of laboratory animals. Procedure for Obtaining Renal Proximal Tubules Before each experiment, a rabbit was anesthetized with an im injection containing both ketamine (7 mg kg ) and xylazine (3 mg kg). Once the rabbit was in a state of deep anesthesia, the abdomen was opened and the kidneys were perfused in situ with 12 ml of a cold (4#{176}C) phosphate-sucrose buffer solution (8) that was Infused into the abdominal aorta in a retrograde manner. After the kidneys were perfused for a few minutes and had fully blanched from the removal of blood, they were removed from the animal. The kidneys were then quickly sliced into 1- mm coronal sections. The sections were stored in the same phosphate-sucrose buffer solution. All segments of the proximal tubule were dissected from these slices under a dissecting microscope for the next 8 to 12 h (8). We dissected Si segments of the proximal tubule from the outer or subcapsular cortex. This segment of the proximal tubule was identified by its large outside diameter and convoluted shape. A renal corpuscle was found on occasion to be attached to some of the dissected Si segments, which provided unequivocal proof of their identity. The S2 segments were straight portions of the proximal tubule that spanned the entire thickness of the cortex. We obtained S3 segments from the outer stripe of the outer medulla. These tubules were Identified as the last 1 mm of the proximal tubule that was attached to the descending thin limb of the loop of Henle. Method for Perfusing Isolated Segments of the Proximal Tubule We transferred each dissected segment of the proximal tubule to a lucite perfusion chamber and suspended it between two sets of pipettes. One set of pipettes was used to perfuse the suspended tubule, and the other was used to collect the perfusion fluid emanating from the lumen of the tubule opposite to the end of the tubule that was being perfused. ach tubule was allowed to warm to 37#{176}Cbefore our observations were started. Perfusion rate was on average maintained at approximately 1 nl min (Table 1) with hydrostatic pressure. The perfused solution was collected in a constant volume pipette (5 nl) to determine the rate at which fluid was being collected. The bathing fluid outside of the suspended tubule was pumped Into the bathing chamber at a rate of.26 ml min and was continually aspirated and collected into scintillation vials at 5-mm intervals. The perfusing chamber contained a total volume of Journal of the American Society of Nephrology 867

3 actors Affecting Renal Hg2 Transport TABL I. Perfusion data#{176} Substances Added To Perfusing luid Segment of Proximal Tubule Length of Segment (mm) Perfusion Rate (ni/mm) Mean Luminal (Hg2) (MM) (Hg2) in Collectate JDJA (fmol min1 mm1) HgCI2 ( (N= 4).54 ± ± ± ± ± ± 1.96 MM) plus S2 (N= 4).88 ± ± ±.81b 11.2 ±.61b 6. ± ± 3.28 Glutathione 53(N= 4).66 ± ± ± 1,3b 8.5 ±.65c III ± 18.7C 7.82 ±.44 (8 MM) H9CI2 (18.4 SI (N= 3).6 ± ± ± ± ± ± 2.44 MM) plus S2 (N= 3) 1.17 ± ± ± ±.18b 61. ± ± 2.32 Cysteine(8 S3(N=3).96±.2 1.8± ± ±O.l#{243} 94.6± ± 1.89 MM) HgCI2 (18.4 SI (N= 3).73 ± ± ± ± ± ± zM)plus 52(N=3) 1.23± ± ± ± ± ± 1.68 Plasma UI- S3 (N= 3) 1.3 ± ± ± ± ± ±.67 trafiltrate O Values are Mean ± S. JD. disappearance flux of Hg2C from the luminal fluid; JA. appearance flux of Hg2C into the bathing solution. Statistical evaluations for data in each perfusion-treatment group were carted out by first performing one-way analysis of variance for three levels. The protected b Significantly tmultiple different comparison (P<.5) test was from the implemented corresponding when mean significant for the SI Values segments were perfused obtained with with the the same analysis solution. of variance. C Significantly different (P<.5) from the corresponding mean for the SI and S2 segments perfused with the same solution. d Significantly different (P<.5) from the corresponding mean for the S2 segments perfused with the same solution..3 ml of fluid. The techniques used to perfuse the segments of the proximal tubule in vivo are based on the modifications outlined by this and other laboratories (1,9-i 1). Solutions The transport (lumen-to-bath) and toxicity of 18.4 MM inorganic mercury in the presence of glutathione (8 MM), L-cysteine (8 MM), rabbit serum albumin (1 MM), and the ultrafiltrate of plasma obtained from rabbits were studied in all three segments of the proximal tubule. The concentrations of glutathione, cystemne, and albumin that were used in this study were chosen because they theoretically provided enough free sulfhydryl groups to bind up all of the mercuric chloride present in the perfusing solution. or molecules like cystemne and glutathione, each mercuric ion can form a coordinate complex with two molecules of cysteine or glutathione. Because each mole of mercuric ions can theoretically bind to 2 mol of cysteine or glutathione, we needed a mole ratio of at least two to one. We increased this ratio to over four to one to insure a higher probability of each mercuric Ion interacting with the free sulfhydryl(s) of glutathione. cystemne, or albumin present in the perfusing solution. The basic solution used to make up the perfusing and bathing solutions was the same in all sets of experiments, except for one where the ultrafiltrate of plasma was perfused through the three segments of the proximal tubule. This basic solution contained the following: Na, 145 mm; Cl-, 14 mm; K, 5 mm; Ca2, 2.5 mm; Mg2, 1.2 mm; SO4, 1.2 mm; HPO42/ H2P42, 2. mm; D-glucose, 1 mm. The final bathing solution contained all of the constituents in the basic electrolyte solution plus.5 mm L-glutamine. This bathing solution was used in all of the experiments. The perfusing solution contained all of the constituents in the basic electrolyte solution plus 18.4 MM inorganic mercury (23HgC12; 8.8 MCi ml, 2.16 mci mg ) and [3H]L-glucose (5 MCi ml ; 58.8 mci mg- ). The radioactive inorganic mercury was obtained from Buffalo Materials Corporation (Buffalo, NY), and the tritiated L-glucose purchased from NN Research Products (Dupont Co., Boston, MA). The tritiated L- glucose was used as a volume marker and a predictor of the leak of mercury across the tubular epithelium. In the experiments where the effects of glutathione, cystemne, or albumin on the transport and toxicity of inorganic mercury were studied, glutathione, cysteine, or albumin was added to the perfusing solution, respectively, along with the 23HgC12 and [3H1L-glucose. The inorganic mercury and glutathione, cysteine, or albumin were added to the perfusing solution at the same time in order to reduce the level of oxidation of free sulfhydryl(s). In the experiments where the effects of plasma ultrafiltrate were studied, the 23HgC12 and [3H]L-glucose were added to the ultrafiltrate, which was perfused through the lumen of the tubules. 868 Volume 2 Number

4 Zalups et al Technique for Obtaining Plasma Ultraf ilt rate from the Rabbit We obtained 4 to 5 ml of blood from the inferior vena cava of several anesthetized rabbits. The blood was placed in a centrifuge to separate the cellular elements from the plasma. About 2 to 3 ml of plasma were obtained from each sample of blood. The plasma was placed in an amicon filtration cell (model 3 ultrafiltration cell; Amicon Division of W.R. Grace and Co., Danvers, MA) designed to filter small samples (1 to 3 ml). The plasma was filtered under pressure through a filtration membrane having a molecular weight cutoff of 5,. The filtered plasma was divided into 5 ML samples and was then frozen at -8#{176}C. Measurement of the Concentration of ree Thiols in The Perfusate The concentration of free sulfhydryl groups (thiols) in the perfusate containing the glutathione, cystemne, albumin, or the ultrafiltrate of plasma was measured with and without the presence of 18.4 MM inorganic mercury. ree sulfhydryl groups in the various perfusates were measured by the colorimetric method of llman and Lysko (12). This method is based on the release of chromophore paranitrophenylate as the result of the interaction between free sulfhydryls and 5,5 -dithio-bis-(nitrobenzoic acid). The release of the chromophore Is proportionate to the concentration of free sulfhydryls. Absorbance was recorded at 41 nm. Assessment of Cellular Injury in Perfused Segments of the Proximal Tubule xclusion of D&C Green Vital Dye and Qualitative Observations. Cellular injury in perfused segments of the proximal tubule was evaluated in part by qualitative observations under an Inverted biological microscope. Based on findings obtained from our previous article (1), the earliest signs of cellular injury in any of the three segments of the proximal tubule are manifested as cellular swelling. As cellular injury becomes more severe, cytoplasmic vacuolization and blebbing of the lummnal membrane begins to occur. Blebbing of the luminal membrane, however, is not always associated with cellular necrosis, for the cells that are losing luminal membrane often retain the ability to exclude the vital dye D&C green. When tubular epithelial cells begin to undergo necrosis (cell death), they begin to take up the vital dye D&C green. At this time, the cells lose their structural integrity. Leak of the Volume Marker i,-glucose. The leak of the volume marker L3HJL-glucose was used as a quantitative Index of epithelial damage in the perfused segments of the proximal tubule. This leak marker was chosen for our studies because it has a low molecular weight and because it is not transported by renal epithelia. When cellular damage is severe or when tubular epithelial cells undergo cellular necrosis, the leak of L-glucose from the luminal fluid to the bath increases markedly (1,13). Measurement of Transport and Accumulation of Inorganic Mercury The transport of inorganic mercury in each perfused segment of the proximal tubule was determined by the disappearance flux (JD) of Inorganic mercury from the luminal fluid into the tubular epithelial cells and by the appearance flux (JA) of inorganic mercury from the tubular epithelial cells into the bath. The content of inorganic mercury that accumulated in the perfused segments was also determined. The accumulated inorganic mercury was separated Into two fractions. One fraction was a trichloroacetic acid (TCA)-precipitable fraction, and the other was a TCAsoluble fraction. The fractionation was performed in order to determine if most of the inorganic mercury that accumulated within the tubular epithelial cells was associated with intracellular proteins. ractionation of Tubular pithelial Cells At the conclusion of each experiment, we harvested the perfused tubule by grabbing it near the perfusion pipette with a pair of fine forceps and rapidly pulling it free from the perfusing and collecting pipettes. The tubular segment was quickly removed from the bathing solution and was placed Into 1 ML of a solution containing 3% TCA. The tubule was allowed to become rigid in shape and opaque in color before it was removed from the solution with a fine glass needle and placed in a scintillation vial for counting. The content of inorganic mercury in the TCA-treated tubular segment was determined isotopically. The 1 ILL of solution containing the 3% TCA was transferred to a scintillation vial. The chamber containing the solution was rinsed three times, and the diluent was placed in the same scintillation vial. The amount of inorganic mercury in the scintillation vial was determined by scintillation counting. Isotopic Counting Methods The activity of 23Hg2 and 3H in each sample was determined in a Beckman LS-581 scintillation counter by standard methods for separating radioisotopic activities. Scintillation vials contained 8 ml of Parkard Opti-luor scintillation fluid and 1.3 ml of water to maintain the same degree of quenching In all samples. Journal of the American Society of Nephrology 869

5 actors Affecting Renal Hg2 Transport Calculations The rate of disappearance (JD; femtomoles per minute per millimeter) of inorganic mercury from the luminal fluid was determined by the equation: J = PHg X (G/G) X (C - CHg)< CI) where PHg is the concentration (femtomoles per nanoliter) of inorganic mercury in the perfusate and CHg is the concentration (femtomoles per nanoliter) of inorganic mercury in the collectate. The rate at which fluid was collected (nanoliters per minute) is designated as C. This rate was calculated from the time it required to fill the constant volume pipette (5 nl). G is the concentration (counts per minute per nanoliter) of the volume marker 13H)L-glucose in the collectate. and G is the concentration (counts per minute per nanoliter) of the volume marker [3H]L-glucose in the perfusate. PHg and C were calculated from the SA of 23Hg (counts per minute per femtomole). We calculated the rate of appearance of inorganic mercury In the bathing solution (JA; femtomoles per minute per millimeter) by the equation: JA = CPM/(SA x T) where CPM is the activity (counts per minute) of 23Hg2 that appeared in the bathing solution in T (time) minutes. Both JD and JA were normalized by the length (millimeters) of the perfused tubule. Cellular volume (CV) for each perfused segment of the proximal tubule was calculated in picoliters from the following equation: CV = ir (OR2-1R2) X L X.7 where OR and IR are the outer and inner tubular radii (micrometers), respectively, L is the length of the tubule (micrometers), and.7 represents the fraction of cellular volume that is water (9). The cellular concentration of inorganic mercury (CC; micromolar) was calculated by the equation: where CPM is the activity of 3H in counts per minute that appeared in the bathing solution in T time (minutes). The nonspecific leak of I3HIL-glucose was normalized by tubular length (millimeters). SA is the specific activity of [3H]L-glucose in terms of counts per minute per nanoliter. The expected leak of inorganic mercury (LHg; femtomoles per minute per millimeter) was calculated by the following equation: LHg = (L X PHg)/L where L is the volume leak (nanoliters per minute), PHg is the concentration of inorganic mercury (femtomoles per nanoliter) in the perfusate. and L is the length (millimeters) of the perfused tubular segment. Leak of (3HJL-glucose (J) measured as femtomoles per minute per millimeter was calculated by the following equation: Statistics JD = Lr, X (G/SA) All values are mean ± S. Significant differences between means for data obtained from the Si, S2, and S3 segments of the proximal tubule that are presented in the figures in this manuscript were determined by first performing a 3 x 2 or 3 x 3 twoway analysis of variance followed by performing the Tukey s protected t multiple comparison test. ach set of data for the three segments of the proximal tubule presented in Table 1 were analyzed with a one-way analysis of variance for three levels (the three segments of the proximal tubule) and then were analyzed with the Tukey s protected t multiple comparison test. The level of significance (a; two tailed) for all analyses was chosen a priori to be.5. RSULTS CC = (CPM/SA)/CV Toxicity xperiments where CPM is the activity (counts per minute) of 23Hg In the cellular extract, SA is the specific activity of 23Hg2 that was calculated from standards (counts per minute per femtomole) and CV is cellular volume. Values for CPM of 23Hg2 were corrected for any 23Hg2 that may have remained In the tubular lumen. This correction was determined from the amount of volume marker (13H]L-glucose) that was extracted with the tubule. The amount of correction needed was normally no greater than 1%. The nonspecific leak (L6; nanoliters per minute per millimeter) measured by the appearance of [3H]L-glucose (volume marker) in the bathing solution was determined by using the following equation (13): L = CPM/(SA x T) Qualitative Observations. When 18.4 MM inorganic mercury is perfused through the lumen of the Si, S2, or S3 segment of the rabbit proximal tubule, cellular necrosis occurs within 5 mm at the end of the tubular segment attached to the perfusion pipette (1). In the study presented here, we discovered that when 8 MM glutathione was added to the perfusate along with the 18.4 MM inorganic mercury, no visible sign of cellular injury occurred in any portion of the Si, S2. or S3 segments of the proximal tubule for up to 1 h of perfusion. In several experiments, the vital dye D&C Green was added to the perfusate to determine whether the absence of cellular necrosis could be substantiated by the exclusion of the dye from the tubular epithelial cells. In all of the perfused tubular segments, there was no evidence of the vital dye 87 Volume 2 Number 4 #{149} 1991

6 Zalups et al being taken up by any of the tubular epithelial cells. Thus, it appears that 8 MM glutathione in the perfusion solution provides complete protection to all segments of the proximal tubule from the toxic effects of 18.4 MM inorganic mercury, when it is perfused through the lumen of the tubular segments. Different toxicological findings were obtained for the three segments of the proximal tubule when 8 MM cysteine was substituted for glutathione in the perfusing solution. Within 1 mm of initiating the perfusion of the electrolyte solution containing the Inorganic mercury and cysteine, generalized swelling accompanied with apical vacuolization occurred in the epithelial cells In the Si segments. In addition, several epithelial cells in some of the perfused Si segments displayed signs of cellular blebbing during the initial 1 mm of perfusion. No further changes occurred In any of the Si segments for up to 45 mm of perfusion. Cellular swelling and vacuolization also occurred in the S2 segments during the initial 1 mm of perfusion, but only in the first 5 to 7 Mm of the perfused tubule. In the S2 segments, cellular blebbing did not occur and there was no progression of cellular injury after the initial 1 mm of perfusion. The S3 segments of the proximal tubule were least affected by the perfused inorganic mercury when cysteine was present in the perfusate. Throughout the entire time the S3 segments were being perfused, they appeared to be free of any pathology, with the exception of one segment. In this segment, slight cellular swelling occurred in the first 5 Mm of the tubule during the initial 1 mm of perfusion. In the experiments where the three segments of the proximal tubule were perfused with the electrolyte solution that contained the inorganic mercury and 1 MM rabbit serum albumin, we observed severe generalized swelling throughout the entire lengths of all three segments of the proximal tubule during the initial 1 mm of perfusion. In each case, the cellular swelling became so severe that the tubular lumen was obstructed to the point that no perfusion fluid could be perfused through the tubule to the collection pipette. It should be pointed out that cellular swelling did not occur in any of the segments of the proximal tubule when they were perfused under conditions where the electrolyte solution did not contain any inorganic mercury but did contain the rabbit albumin. These findings indicate that the pathological changes that occurred were not induced by the albumin alone. We also observed some pathological changes in a couple of the segments of the proximal tubule that were perfused with the ultrafiltrate of rabbit plasma that contained the 18.4 MM inorganic mercury. In the Si segment, cellular swelling occurred throughout the entire tubule by the end of the first 15 mm of perfusion. The swelling was not as bad, however, as that which occurred in the experiments where albumin was added to the perfusate. We were able to collect tubular fluid in these experiments. Interestingly, no pathological changes were observed in any of the S2 segments that were perfused with the plasma ultrafiltrate containing 18.4 MM inorganic mercury. Some pathological changes were observed In the S3 segments by the end of first 15 mm of perfusion. A moderate degree of cellular swelling occurred In the Initial 5 Mm of the tubule. Occasional blebbing occurred in several of the epithelial cells in some of the tubules. Notwithstanding, there was no progression in the severity of any of the pathological changes after the first 15 mm of perfusion. Leak of Volume Marker. igure 1 indicates the leak of the volume marker 3HJL-glucose (femtomoles per minute per millimeter) across the tubular epithehum of isolated Si, S2, and S3 segments of the proximal tubule perfused with 18.4 MM inorganic mercury in the presence of 8 MM glutathione, 8 MM cystelne, or the ultrafiltrate of rabbit plasma. According to the leak of the I3HJL-glucose, cysteine provided the lowest level of protection against the toxic effects of 18.4 MM inorganic mercury, particularly in the 51 segment. This is consistent with the visual observations that the toxic effects of the 18.4 MM inorganic mercury were greatest in the Si segments perfused with 8 MM cysteine. lux xperiments Some of the relevant statistics from the flux experiments are presented in Table 1. In addition, the concentrations of free sulfhydryl groups in the presence and absence of 18.4 MM inorganic mercury in the various perfusates used in this study are presented in Table 2. In general, the concentration of free sulfhydryl groups was decreased by approximately 5% when 18.4 MM inorganic mercury was added to any of the perfusates. including the plasma ultrafiltrate. JD and JA of Inorganic Mercury. The magnitude of the JD (femtomoles per minute per millimeter) of inorganic mercury from the tubular fluid and the magnitude of the JA (femtomoles per minute per millimeter) of inorganic mercury into the bathing solution for the three segments of the proximal tubule under conditions where inorganic mercury (18.4 MM) was perfused, along wtih glutathione (8 MM) or cystelne (8 MM), or in plasma ultrafiltrate are presented in igure 2. The J of inorganic mercury Increased markedly from the Si to the S3 segment of the proximal tubule when glutathione was present in the perfusate. In the 51 segments, the JD of inorganic mercury was 24.4 ± 2.46 (mean ± S). This value more than doubled in the S2 segments (6. ± 5.18) and more Journal of the American Society of Nephrology 871

7 actors Affecting Renal Hg2 Transport Ui (3 U, - ọ.i u-. Ui t Giutathione (8 pm) Cysteine (8 pm) Ultrafiltrate 53 SGMNTS O TH PROXIMAL TUBUL igure 1. The leak of the volume marker (3H)i-glucose (ferntomoles per minute per millimeter) across the tubular epithellum of Isolated SI, S2, and S3 segments of the proximal tubule perfused with 18.4MM inorganic mercury in the presence of 8 MM glutathione (N = 4 for SI, $2, and for S3 segments), 8 MM cysteine (N = 3 for SI, S2, and S3 segments), or the ultrafiltrate (N= 3 for SI, S2. and S3 segments) from rabbit plasma. Values ore mean ± S. Significontly different (P<.5) from the mean for the corresponding SI segments perfused with either 8 MM glutathione or the ultrofiltrote of rabbit plasma. tsignlflcantly different (P <.5) from the mean for the SI segments perfused with 8 MM cystemne. than quadrupled in the S3 segments (iii ± 18.7). By contrast, the JA of inorganic mercury was statistically similar in all three segments of the proximal tubule when glutathione was added to the perfusate containing 18.4 MM inorganic mercury. The level of JA in the Si (1.6 ± 1.96), S2 (15.5 ± 3.28), and S3 (7.82 ±.44) segments of the nephron was low. When compared with the Jo of inorganic mercury, the JA of inorganic mercury was significantly greater (P <.5) in the S2 and S3 segments, but not in the Si segments. When cystemne was added to the perfusate, inorganic mercury moved rapidly from the luminal fluid into the tubular epithelial cells in all three segments of the proximal tubule. Interestingly. the J of inorganic mercury in the Si segments (145 ± 52.8) was significantly greater than that In the S2 segments (61. ± 2.16) and was statistically similar to that in the S3 segments (94.6 ± 11.9). The opposite occurred when ghutathione was added to the perfusion solution. In this situation, the J of inorganic mercury was lowest in the Si segments. In all three segments of the proximal tubule, the JD of Inorganic mercury was significantly greater than the JA of inorganic mercury. The movement of inorganic mercury from the tubular epithehial cells to the bath, as measured by the JA of inorganic mercury, was low In the Si (14.8 ± 2.44), S2 (7.87 ± 2.32), and S3 (8.88 ± 1.89) TABL 2. Concentrations of free sulfhydryl groups in perfusates in the presence and absence of 18.4MM inorganic mercury#{176} Perfusote Concentration of Concentration of ree SH-Groups ree SH-Groups In The Absence In The Presence of I8.4MHg2 of 18.4 MM Hg2 (MM) lectrolyte Solution 8 ± 5 46 ± 5 Plus Glutothione N = 3 N = 3 (8 MM) lectrolyte Solution 1 ± ± 8 Plus Cysteine (8 N= 3 N- 3 MM) lectrolyte Solution 13 ± 1 44 ± 13 Plus Plasma Albu- N= 3 N= 3 mm (1 MM) Ultrafiltrate of Rabbit 59 ± 8 26 ± 6 Plasma N=3 N=3 #{176} are mean ± S. The composition of the electrolyte solution Is described in the text. The inorganic mercury and glutathione. cysteine, or plasma albumin were added to the electrolyte perfusing solution simultaneously. segments of the proximal tubule. Moreover, there was no significant difference between the three segments with respect to the mean value of the JA of inorganic mercury. When we perfused inorganic mercury in the ultrafiltrate of rabbit plasma, the J of inorganic mercury in the Si (46.8 ± 6.48), S2 (41.2 ± 12.9), and S3(46.1 ± 4.3) segments was somewhat low. No significant difference was found between any of the three segments with respect to the J of inorganic mercury. Although the JD of inorganic mercury was low in each segment, it was significantly greater than the corresponding JA of inorganic mercury, which was very low. No significant differences were found between the values for the JA of inorganic mercury in the Si (7. ± 2.73), S2 (4.25 ± 1.68), and S3 (6.51 ±.67) segments of the proximal tubule. Tubular Leak and Transport of Inorganic Mercury. The predicted leak of inorganic mercury (LHg; femtomoles per minute per millimeter), from the luminal fluid to the bathing solution was small In the Si (3.82 ±.4i), S2 (4.62 ± 1.6), and S3 (4.25 ± 1.12) segments of the proximal tubule when 8 MM glutathione was present in the perfusate (igure 3). In addition, there is no significant difference In the magnitude of leak between the three segments. In all three segments, the actual measured flux of Inorganic mercury into the bath (JA) exceeded L. Statistically significant differences between JA and LHg, however, were found only in the Si and S2 segments. On the basis of these data, it appears that some 872 Volume 2 Number

8 Zalups et ai C, I I LI_ Glutathione 77 Disappearance lux IJD) Appearance In Bath lux (J) cysteine * I I- in I- z at = ) Giutathione U- 1 ±1 >( -J U- 5 Ultra tilt r ate 15 I.Ultra filtrate * I I Predicted Leak into Bath Actual lux Measured cysteine di 6 3 LLTL SI S2 S3 SGMNTS O TH PROXIMAL TUBUL igure 2. The JD (femtomoles per minute per millimeter) of inorganic mercury (Hg2) from the luminal fluid of, and the JA (femtomoles per minute per millimeter) of inorganic mercury in the bath from, SI, S2, and $3 segments of the proximal tubule when Hg2 (18.4 MM) was perfused through the lumen of these segments in the presence of glutathione (8 MM; N = 4 for SI, S2, and S3 segments), cysteine (8 MM; N = 3 for SI, S2, and S3 segments), or the ultrafiltrote of rabbit plasma (N = 3 for SI, 52, and S3 segments). The values ore mean ± S. Significantly different (P<.5) from the corresponding mean for JA of Hg2. Significantly different (P <.5) from the corresponding mean for JA of Hg2 and the mean for the J of Hg2 for the S2 segment of the proximal tubule under the same perfusing conditions. tsig nificantly different (P <.5) from the mean for the JD of Hg2 for the SI segment of the proximal tubule. transepithelial transport of inorganic mercury occurred in the proximal tubular epithelium when glutathione was present in the perfusate, specifically in the Si and S2 segments. B C B ) 1 5 I SI S2 S3 SGMNTS O TH PROXIMAL TUBUL igure 3. The predicted leak (L. femtomoles per minute per millimeter) and actual JA (femtomoles per minute per millimeter) of inorganic mercury (Hg2) in the both of SI, 52, and S3 segments of the proximal tubule when Hg2 (18.4 MM) was perfused through the lumen of these segments in the presence of glutathione (8 MM; N = 4 for SI, S2, and 53 segments), cysteine (8 MM; N = 3 for SI, S2, and S3 segments), or the ultrafiltrate of rabbit plasma (N= 3 for Si, S2, and S3 segments). The values are mean ± S. SIgnlficantly different (P<.5) from the mean for the corresponding of Hg2. tsignificontly different (P <.5) from the mean for the predicted leak of Hg2 for the SI segment of the proximal tubule under the same perfusing conditions. In those experiments where 8 MM cystemne was added to the perfusate, LHg was very large In the Si segments (9.3 ± 2.3). In fact, LHg was significantly greater than the JA of inorganic mercury. L in the S2 (1.3 ± 5.16) and S3 (1.2 ± 5.3) segments was Journal of the American Society of Nephrology 873

9 actors Affecting Renal Hg2 Transport significantly lower than that in the Si segment. Moreover, no significant difference between the means for L and the JA of inorganic mercury were calculated for either the Si or S2 segments. Thus, the amount of mercury that entered the bath in these segments can be attributed to the leak of mercury across the tubular epithelium. When inorganic mercury (18.4 MM) was perfused in the ultrauiltrate of rabbit plasma, the L in the Si (4.4 ±.94), S2 (3.67 ±.94), and S3 (7.73 ± 4.7) segments was small, and it was not significantly different from the JA of inorganic mercury. This indicates that all of the inorganic mercury that entered the bath from each of the three segments of the proximal tubule can be explained by the leak of inorganic mercury across the tubular epithelium. Inorganic Mercury Accumulated Within the Thbular pithelluni. At the end of the perfusion experiments in which inorganic mercury (18.4 MM) and glutathione (8 MM) were present in the perfusing solution, more than twice as much inorganic mercury (picomoles per millimeter) was present in the TCAprecipitable fraction of the S3 segments (1.8 ±.2) than in the TCA-precipitable fraction of the Si (.72 ±.43) or S2 (.64 ±.17) segments (igure 4). There was no significant difference in the content of inorganic mercury in the TCA-precipitable fraction of the tubular epithelium between the Si and S2 segments. Very little mercury was detected in the TCA-soluble fraction of the Si or S2 segments. Only in the TCAsoluble fraction of the S3 segment was there an appreciable amount of inorganic mercury (.16 ±.2). When cysteine was present In the perfusing solution, the accumulation of inorganic mercury along the proximal tubule was opposite of that when glutathione was in the perfusing solution. More than twice as much Inorganic mercury was present in the TCA-precipitable fraction of the Si segments (1.22 ±.19) than in the TCA-precipitable fraction of either the S2 (.41 ±.1) or S3 (.52 ±.7) segments. The content of inorganic mercury in the TCA-preclpitable fraction of the S2 and S3 segments was not significantly different. ssentially no mercury was present in the TCA-soluble fraction of any of the three segments of the proximal tubule. The content of inorganic mercury in the Si (.13 ±.3), S2 (.38 ±.13), and S3 (.26 ±.5) segments of the proximal tubule was comparatively low in the experiments where inorganic mercury was perfused In the ultrafiltrate of rabbit plasma. There was no significant difference between any of the three segments with respect to the content of inorganic mercury in the TCA-preclpitable fraction of the tubular epithelium. No inorganic mercury was found in the TCA-soluble fraction of any of the three segments of the proximal tubule. LU -J in B 4, I- z : B I- z LU I- z () 4, a, 2. * * I Glutathione TCA Precipitable raction TCA Soluble raction cysteine rlfl * Iiltrafiltrate SGMNTS O TH PROXIMAL TUBUL igure 4. The content (picomoles per millimeter) of inorganic mercury (Hg2) in the TCA-precipitabie fraction and the TCA-soluble fraction of SI, S2, and S3 segments of the proximal tubule after the segments were perfused through the lumen with Hg2 (18.4 MM) in the presence of glutathione (8 MM; N= 4forSI, S2, and S3 segments), cysteine(8 MM; N = 3 for SI, S2. and S3 segments), or the ultrafiltrate of rabbit plasma (N = 3 for SI, S2, and S3 segments). The values are mean ± S. Significantly different (P<.5) from the corresponding mean for the content of Hg2 in the TCAsoluble fraction of the same segment. tsigniflcantly different (P<.5) from the mean for the content of Hg2 in the TCAprecipitable fraction of the SI segment of the proximal tubule. ttsmgnificantly different (P <.5) from the mean for the content of Hg2 in the TCA-precipitable fraction of SI and S2 segments of the proximal tubule. S2 S3 874 Volume 2 Number

10 Zalups et al DISCUSSION We recently demonstrated that inorganic mercury (Hg2 ) has pronounced acute toxic effects in isolated Si, S2, and S3 segments of the proximal tubule when it is perfused through the lumen at a concentration of 18.4 MM (i). In these experiments, cellular injury and necrosis occurred rapidly at the end of each tubular segment that was attached to the perfusing pipette. The toxic effects of inorganic mercury, however, were rarely seen beyond the first half of the perfused tubules. We postulated that cellular injury did not occur beyond the first half of each perfused tubule because of a complex ligand interaction preventing the perfused inorganic mercury from either entering into, or adversely affecting, the tubular ep- Ithelial cells distal to the injured region of the perfused tubular segment. Indirect evidence led us to infer that the perfused inorganic mercury interacted with and became bound to cellular biomolecules that were being released into the luminal fluid from injured and necrotic tubular epithelial cells that were near the perfusing pipette. The most likely ligands that the perfused inorganic mercury interacted with were cellular peptides, proteins, and/or amino acids containing free (reduced) sulfhydryl groups. This conclusion Is based on the fact that mercury in plasma is mainly bound to the sulfhydryl groups of plasma proteins, particularly albumin (4,5). It should be pointed out that no compounds containing free sulfhydryl groups were present in the perfusing solution, which consisted mainly of electrolytes. One of the aims of the study presented here was to test whether certain blomolecules containing free sulfhydryl groups, such as the amino acid cysteine, the small peptide glutathione, or the protein albumin, afford any in vitro protection to isolated perfused segments of the proximal tubule from the toxic effects of inorganic mercury. The effect of an ultraflltrate of plasma from the rabbit was also tested because the constituents in the ultrafiltrate presumably are similar to those present in the lummnal milieu of the proximal tubule in vivo. We observed that cystemne (8 MM), glutathione (8 MM), and the ultrafiltrate of rabbit plasma provided various degrees of protection to Si, S2, and S3 segments of the proximal tubule in vitro when they were perfused through the lumen of the tubular segments concomitantly with inorganic mercury (18.4 MM). Albumin (1 MM), on the other hand, provided no protection against the cytotoxic effects of inorganic mercury when it was added to the perfusate. It should be mentioned that a more than four times greater molar concentration of cysteine, glutathione. and albumin were used in the perfusate in order to provide enough ligands to potentially bind all the divalent atoms of mercury. Adding glutathione to the perfusate provided the epithelial cells in all three perfused segments of the proximal tubule complete protection from the cytotoxic effects of inorganic mercury. No evidence of cellular injury was observed in any of the three perfused segments of the proximal tubule. The ultrafiltrate of plasma provided the second greatest protection. However, complete protection was not afforded to all segments of the proximal tubule as was the case with glutathione. Some cellular injury was observed in the Si and S3 segments, particularly in the Si segment. The level of protection provided by cysteine was less than that provided by the plasma ultrafiltrate. Cellular injury occurred mainly in Si and S2 segments of the proximal tubule, but the level of injury was less than that observed with inorganic mercury alone, The assessment of tubular leak of the volume marker 3HJL-glucose revealed that only Si segments that were perfused with 8 MM cystemne were injured severely enough to cause a significant increase in the intercellular and/or transcellular movement of the volume marker. Because the volume marker [3HJLglucose is not transported by renal epithelia, the increased leak of this marker in the Si segments was due to either increased permeability in the intercellular junctional complexes or actual movement of the volume marker through necrotic epithelial cells. When albumin was present in the perfusate, inorganic mercury caused swelling and cellular injury along the entire length of all three of the perfused segments of the proximal tubule. This is in sharp contrast to what happens when inorganic mercury is perfused through the same segments of the proximal tubule In the absence of albumin. Thus, albumin appears to somehow carry or spread the toxic effects of inorganic mercury along the entire length of perfused segments of the proximal tubule. urther studies are required to determine the interactions between albumin, inorganic mercury, and the tubular epithelial cells along the entire length of the proximal tubule. There appears to be a common feature associated with the protection afforded by cysteine, glutathione. and the plasma ultrafiltrate. In each case, the cellular uptake of inorganic mercury in the three segments of the proximal tubule was reduced substantially when compared with that measured in experiments where inorganic mercury was not perfused in the presence of cysteine, glutathione. or plasma ultrafiltrate. The cellular uptake of inorganic mercury varied between.2 to 1.7 pmol/mm length of tubule (igure 3). These values are considerably lower than those obtained in our previous experiments (i). In those experiments, when 18.4 MM inorganic mercury was perfused in the absence of any compounds containing free sulfhydryl groups, the cellular uptake of inorganic mercury in Si and S3 segments of the Journal of the American Society of Nephrology 875

11 actors Affecting Renal Hg2 Transport proximal tubule was 3.5 and 3.1 pmol/mm, respectively (i). Thus, we can infer that some of the protection afforded to the segments of the proximal tubule by cysteine, glutathione, and plasma ultrafiltrate may be related to reduced entry of inorganic mercury into the tubular epithelial cells. The mechanism for the reduced entry of inorganic mercury into the tubular epithelial cells is not known at present. However, knowing some of the binding characteristic of inorganic mercury (i 4,15), we suspect that inorganic mercury forms linear two coordination complexes with cysteine, glutathione, and small peptides (In the ultrafiltrate) in the perfusate. indings from [ 3C] NMR experiments with glutathione and inorganic mercury indicate that when glutathione and Hg2 are In a ratio of 2:1 in solutions having a ph range between less than 1 to greater than ii, a 2:1 complex of glutathione and Hg2 forms with each atom of mercury binding exclusively to the sulfhydryl group of the cysteinyl moiety of two glutathione molecules (15,16). indings from the study presented here are consistent with the findings from the NMR experiments. When 18.4 MM inorganic mercury was added to any of the perfusates. It caused a 5% reduction in the concentration of free sulfhydryls in the perfusates. Because the mole ratio of free sulfhydryls to inorganic mercury was approximately 4:1 in the perfusates containing either glutathione or cysteine, a reduction of this ratio to about 2:1 is consistent with the concept of each mercuric ion forming a coordinate complex with two molecules of glutathmone or cystemne. It is possible that once these complexes are formed, they do not pass into the tubular epithelial cells as readily as does free ionic mercury. Thus, there is less accumulation of inorganic mercury into the tubular epithelial cells and less cellular injury. Our present findings show that some compounds containing free sulfhydryl groups interact with inorganic mercury in a manner that prevents inorganic mercury from entering into the proximal tubular epithelium from the lumen. These findings support our theory as to why the distal portions of perfused segments of the proximal tubule were protected from the toxic effects of inorganic mercury when inorganic mercury was perfused in the absence of compounds containing free sulfhydryl groups. Specifically, the findings support the theory that cellular injury is prevented downstream of injured proximal tubular epithelial cells by the inhibition or prevention of the entry of inorganic mercury into the distally situated epithelmal cells as a result of the binding of inorganic mercury to the sulfhydryl groups of molecules released from injured and necrotic epithellal cells more proximally. Reduced cellular uptake of Inorganic mercury may only partially explain the complete protection provided to segments of the proximal tubule by glutathione. Despite the absence of tubular injury when glutathione was present In the perfusate, a substantial amount of inorganic mercury disappeared from the lummnal fluid in the three segments of the proximal tubule, particularly in the S3 segment. Moreover, the levels of JD of inorganic mercury in the S2 and S3 segments was similar when glutathione or cysteine was present in the perfusate (igure 2). However, cellular injury was observed in the S2 segments only when cystemne, and not glutathione, was present in the perfusate. The cellular accumulation of inorganic mercury was also substantial in the three segments of the proximal tubule when glutathione was present in the perfusate, particularly in the S3 segment (igure 3). rom these findings, we can infer that at least some of the inorganic mercury that entered into the tubular epithelium was chemically altered in some manner to prevent cellular injury from being induced. Perhaps some of the inorganic mercury that entered the tubular epithelial cells entered as an inorganic mercury-glutathione complex (glutathione- Hg-glutathione). Some of the inorganic mercury may also have entered in some other state and then formed a complex with glutathione in the cytosol. Such a complex may not readily dissociate within the epithelial cells and thus would prevent the mercuric ion from reacting or binding to intracellular sites that would lead to cellular injury and death. We emphasize that the interactions of inorganic mercury with nonprotein thiols, such as glutathione and metallothionemn in the cytosol, are very complex (15). In order to gain a better understanding of the possible factors involved in the expression and prevention of toxic injury produced by the mercuric ion, the interactions of inorganic mercury with nonprotein thiols and other compounds containing free sulfhydryl groups in the cytosol need to be studied. The factors responsible for the protection afforded by the ultrafiltrate of rabbit plasma cannot be Identified on the basis of the findings from the study presented here. It is likely, however, that some of the protection Is associated with the binding of inorganic mercury to the sulfhydryl groups of amino acids, peptides, and proteins present in the ultrafiltrate. Some of these compounds include cysteine and glutathione. It is not likely, on the other hand, that the glutathione or cysteine in the ultrafiltrate could bind up all the inorganic mercury in the ultrafiltrate. The concentration of glutathione and cysteine In the plasma of arterial blood Is normally about 1 MM (17). Because the concentration of inorganic mercury in the perfusate was 18.4 MM and because inorganic mercury generally forms a coordinate bond with two sulfhydryl groups within the normal range of ph (14). there was, theoretically, not enough glutathione or cysteine to saturate the binding sites of all of the Inorganic mercury present in the perfusate. Because 876 Volume 2 Number

12 Zalups et al of the diversity of proteins present in plasma, it is difficult to make any conclusion about all of the ligands with which inorganic mercury Interacts in the ultrafiltrate of plasma. Only further studies can provide more information about the possible ligands in plasma with which inorganic mercury can interact. The transport of inorganic mercury in the three segments of the proximal tubule varied considerably depending on the constituents present in the perfusate. The greatest degree of variation observed in the study presented here was in the magnitude of JD of mercury from the lummnal fluid (igures 2 and 3). In our previous study (1), we determined that the JD of mercury averaged approximately 15 fmol mmn mm length of tubule In all three segments of the proximal tubule when 18.4 MM inorganic mercury was perfused In the absence of any compounds containing free sulfhydryl groups. or the purpose of comparison, the JD, JA, and the leak data from our previous study are summarized graphically in igure 5. WIth the exception of one case, the JD of inorganic mercury in the three segments of the proximal tubule that was measured in this study was substantially less than 15 fmol min mm length of tubule. Only in the Si segments that were perfused with 18.4 MM inorganic mercury and 8 MM cysteine was the J similar to that seen in any of the segments of the proximal tubule perfused with 18.4 MM inorganic mercury alone. The JA of inorganic mercury into the bath from all of the segments of the proximal tubule perfused with the inorganic mercury in the presence cystemne, glutathione, or plasma ultrafiltrate was very low (igure 2). In every case but one, the JD was significantly greater than the JA. The only exception was when Si segments were perfused with the solution containing the inorganic mercury and glutathione. In this case, the JD and JA of inorganic mercury were both low (less than 3 fmol min mm length of tubule). The inorganic mercury that appeared in the bathing solution from each of three segments of the proximal tubule perfused with the plasma ultrafiltrate or the perfusing solution containing cystemne can be explained by the nonspecific leak of mercury (igure 3). Injury to the tubular epithelium probably contributed to some of the leak. In contrast, when glutathione was present in the perfusate. the rate of appearance of inorganic mercury in the bath exceeded the predicted rate of appearance of inorganic mercury due to nonspecific leak. This was particularly true in the Si and S2 segments of the proximal tubule. The difference in the S3 segment of the proximal tubule was not statistically significant. These data indicate that a small amount (5 to 1 fmol min mm length of the tubule) of transepithelial transport of inorganic mercury occurs in the proximal tubule when gluta * C >(. 1 D 5 25 #{149} I No ree SH-Groupu I,, Pert user. Si :i: S2 jd JA L,5 SGMNTS O TH PROXIMAL TUBUL igure 5. The J, JA, and leak of inorganic mercury in SI, $2, and S3 segments of the proximal tubule perfused with 18.4 MM inorganic mercury in the absence of any compounds containing free sulfhydryl groups. The values are mean ± S. Significantly different (P <.5) from the mean for the corresponding J. The data for this figure were taken from our previous article (1). thione is present in the perfusate. Our previous findings showed that a significant level of transepithelial transport of inorganic mercury occurs in the S3 segment of the proximal tubule when inorganic mercury is perfused in the absence of compounds containing free sulfhydryl groups (igure 5). Thus, the chemical interactions between inorganic mercury and compounds containing free sulfhydryl groups can greatly alter the transport of inorganic mercury in the proximal tubule. When rats or rabbits are given a nontoxic dose of inorganic mercury, about 5% of the dose of inorganic mercury is present in the kidneys within 48 h (2,18-21). Most of the inorganic mercury in the kidneys is located in the cortex and outer stripe of the outer medulla. Virtually all of the inorganic mercury that is in these two zones of the kidney is present in the proximal tubule (2,22). Data from hlstochemical and isolated tubular studies indicate that all segments of the proximal tubule accumulate inorganic mercury (2,22). Our findings presented here confirm the fact that inorganic mercury Is accumulated in all segments of the proximal tubule under a variety of conditions, particularly when the mercury is perfused with an ultrafiltrate of plasma, which presumably closely reflects the conditions found in vivo. What is perplexing, however, is that the pars recta (which includes the last part of the S2 segment and the entire S3 segment [23J) is the portion of the proximal tubule that is primarily affected by the toxic effects of mercury (2,22,24-26). Unfortunately, we do not know why the pars recta of the proximal S3 Journal of the American Society of Nephrology 877

13 actors Affecting Renal Hg2* Transport tubule is the portion of the proximal tubule that is Injured by inorganic mercury. In summary, the findings presented here indicate that cysteine. glutathione, albumin, and the ultrafiltrate of rabbit plasma significantly alter both the transport and toxicity of inorganic mercury in isolated perfused segments of the proximal tubule. Partial to complete protection against the toxic effects of Inorganic mercury is afforded to the three segments of the proximal tubule when the inorganic mercury is perfused in the presence of cysteine, the plasma ultrafiltrate, or glutathione. Moreover, the protection that is afforded appears to be related in part to a diminished entry of inorganic mercury into the tubular epithelial cells in the three segments of the proximal tubule. ACKNOWLDGMNTS The studies described in this article were supported In part by Research Grant S 5157 from the NIH. (R.K.Z.). We gratefully acknowledge Dr. Lawrence H. Lash of the department of pharmacology at Wayne State University for his assistance In determining the concentration of freesulfhydryl groups In samples of the various perfusates used In this study. RRNCS 1. Barfuss DW, Robinson MK, Zaiups RK: Inorganic mercury transport in the proximal tubule of the rabbit. J Am Soc Nephrol i99;i:9i Zalups RK, Barfuss DW: Accumulation of inorganic mercury along the renal proximal tubule of the rabbit. Toxicol Appl Pharmacol 199; 16: Berlin M, Gibson S: Renal uptake, excretion. and retention of mercury. I. A study in the rabbit during infusion of mercuric chloride. Arch nviron Health 1963:6: riedman HL: Relationship between chemical structure and biological activity In mercurial compounds. Ann NY Acad Sd 1957;65: Mussini : Bonds of mercurial diuretics to blood proteins. Boll Soc Ital Biol Sper 1958:34: Huitman P. nestrom S, Von Schenck H: Renal handling of inorganic mercury In mice. Virchows Arch B 1985:49: Hultman P. nestrom S: Localization of mercury in the kidney during experimental acute tubular necrosis studied by the cytochemical silver amplification method. Br J xp Pathol 1986:67: Pine SC, Potts DJ: A comparison of the relative effectiveness of three transplant preservation fluids upon the integrity and function of rabbit proximal convoluted tubules perfused In vitro. Clin Sd 1986:7: Barfuss DW, Schafer JA: Active amino acid absorption by proximal convoluted and proximal straight tubules. Am J Physiol 1979;236: Banfuss DW, Schafer JA: low dependence of nonelectrolyte absorption in the nephron. Am J Physiol l979;236:i ii. Tune BM, Burg MB: Glucose transport by proximal renal tubules. Am J Physiol 1971:221: llman G, Lysko H: A precise method for the determination of whole blood and plasma sulfhydryl group. Anal Biochem i979;93: Barfuss DW, Schafer JA: Difference in active and passive glucose transport along the proximal nephron. Am J Physiol 1981 ;24: Rabenstein DL: Aqueous-solution chemistry of methyl mercury and its complexes. Accts Chem Res i978;ii:i-i Rabenstein DL. Metal complexes of glutathione and their biological significance. In: Dolphin D, Auramovlbc. Poulson R, eds. Glutathione: Chemical, Biochemical, and Medical Aspects. Vol. 3. Coenzymes and Cofactors. New York: Wiley; 1989: uhr BJ, Rabenstein DL: Nuclear magnetic-resonance studies of solution chemistry of metalcomplexes. 9. Binding of cadmium, zinc, lead and mercury by lutathione. J Am Chem Soc 1 973;95: Lash LH, Jones DP: Distribution of oxidized and reduced forms of glutathione and cysteine in rat plasma. Arch Biochem Biophys 1 985;24: Rothstein A, Hayes AD: The metabolism of mercury in the rat studies by isotope techniques. J Pharmacol xp Ther 196:13: ZalupsRK, DiamondGL: Intrarenal distribution of mercury in the rat: ffect of administered dose of mercuric chloride. Bull nviron Contam Toxicol 1987;38: Zalups RK, Diamond GL: Mercuric chloride-induced nephrotoxicity in the rat following unilateral nephrectomy and compensatory renal growth. Virchows Arch B 1987;53: Zalups RK, Kotzbach JM, Diamond GL: nhanced accumulation of injected inorganic mercury in the renal outer medulla after unilateral nephrectomy. Toxicol Appl Pharmacol 1987; 89: Zalups RK: Autometallographic localization of inorganic mercury in the icidneys of rats: ffect of unilateral nephrectomy and compensatory renal growth. xp Mol Pathol 199 i. in press. 23. Kriz W: A standard nomenclature for structures of the kidney. Am J Physiol 1988;23:i Gritzka TL, Trump B: Renal tubular lesions caused by mercuric chloride: lectron microscopic observations: Degeneration of pars recta. Am J Pathol 1968:52: Ganote C, Refiner KA, Jennings RB: Acute mercuric chloride nephrotoxicity: An electron microscopic and metabolic study. Lab Invest 1974:31: McDowell M, Nagie RB, Zahne RC, McNeil JS, lamenbaum W, Trump B: Studies on the pathophysiology of acute renal failure. I. Correlation of ultrastructure and function in the proximal tubule of the rat following administration of mercuric chloride. Virchows Arch B Cell Pathol 1976;22:173-i Volume 2 Number