Plasma cadmium–metallothionein, a biological exposure index for cadmium-induced renal dysfunction, based on the mechanism of its action

Toxicology 129 (1998) 157 – 168

Plasma cadmium – metallothionein, a biological exposure index for cadmium-induced renal dysfunction, based on the mechanism of its action Kazuo Nomiyama *, Hiroko Nomiyama, Naoki Kameda Department of En6ironmental Health, Jichi Medical School, Tochigi-ken 329 -0498, Japan Received 18 February 1998; accepted 9 June 1998

Abstract Thirteen rabbits were given subcutaneous cadmium (0.3 mg Cd/kg) daily. The plasma cadmium – metallothionein (CdMT) and the Cd-induced hepatic and renal functions were determined at 0, 5, 8, 11, 12, 13 and 14 weeks. Hepatic dysfunction, an elevated plasma CdMT and renal dysfunction were detected mostly between 12 and 14 weeks. The hepatic dysfunction parameters were closely related with the plasma CdMT, which was then found to correlate with the renal dysfunction parameters. All the above findings suggest the following mechanism for the Cd-induced renal dysfunction: hepatic CdMT is released into the plasma upon the Cd-induced hepatic dysfunction, and then excess plasma CdMT, whose concentration is proportional to the CdMT in the renal proximal tubular lumen, induces renal dysfunction. The critical concentration of plasma CdMT to induce renal dysfunction was estimated as 80 mg Cd/l. The plasma CdMT is proposed therefore as a biological exposure index for the Cd-induced renal dysfunction, based on the mechanism of its action. © 1998 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Cadmium; Metallothionein; Kidney; Plasma; Critical concentration

1. Introduction It has been suggested that both human and animals chronically exposed to excess cadmium Abbre6iations: Cd, cadmium; CdMT, cadmium–metallothionein. * Corresponding author. Tel.: + 81 285 44 2111, ext. 3137; fax: + 81 285 44 8465; e-mail: nomiyama@jichi ac.jp

(Cd) suffer from renal dysfunction such as proteinuria, glucosuria and aminoaciduria when the level of Cd in the renal cortex exceeds 200 mg Cd/g (World Health Organization, 1991). However, the critical concentration varied with the dose of Cd such as 470, 635 and 1170 mg/g in monkeys fed pelleted food containing Cd at dose levels of 300, 100 and 30 mg Cd/g over periods of 1–9 years (Nomiyama et al., 1979, 1987b). The

0300-483X/98/$19.00 © 1998 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 3 0 0 - 4 8 3 X ( 9 8 ) 0 0 0 8 3 - 3

158

K. Nomiyama et al. / Toxicology 129 (1998) 157–168

critical concentrations of cadmium in the renal cortex of rabbits were 222, 296 and 275 mg Cd/g in the 3, 1 and 0.3 mg Cd/kg groups, respectively, and a critical concentration of the 3 mg Cd/kg group was significantly lower than those of the 0.3 and 1 mg Cd/kg groups at 0.1% level of statistical significance (Nomiyama and Nomiyama, unpublished data). The critical concentration thus decreased with the dose level. Further, the critical concentration of Cd in the renal cortex in rabbits given a single intra-arterial injection of cadmium – metallothionein (CdMT) indicated the critical concentration as low as 13 mg Cd/g (Nomiyama and Nomiyama, 1986). We have the following working hypothesis on the mechanism of Cd-induced renal dysfunction: CdMT, accumulated in the liver following Cd exposure, was released into the blood stream upon the Cd-induced necrosis/apoptosis of hepatic cells, which was caused by excess exposure to Cd, excess intake of alcohol, virus infection or others (Nomiyama and Nomiyama, 1994). Pasma CdMT then passes through glomeruli and produces dysfunction of proximal tubular membrane (Nomiyama and Nomiyama, 1994). To support this hypothesis, we provide the following facts in this studty: (1) the order of overt Cd effects is hepatic dysfunction, an elevated CdMT level in the tubular lumen, and then renal dysfunction, or all effects at the same time; (2) a close relationship exists between hepatic dysfunction and CdMT in the tubular lumen; and (3) a close relationship should also be found between CdMT in the tubular lumen and renal dysfunction.

2. Materials and methods Thirteen male rabbits of the Japanese White strain, weighing 2.5 kg, were housed in stainless steel cages measuring 39 cm in height, 34 cm in width and 47 cm in depth under conditions of controlled temperature (22 92°C) and humidity (5595%), and were maintained on a 12-h light/ dark cycle. The investigation was approved by the Animal Experiment Committee of Jichi Medical School, and all procedures on the animals were carried out according to our institutional guideli-

nes, which fitted the humane guidelines of the American Association of Laboratory Animal Sciences. All 13 rabbits were exposed to cadmium, because the purpose of this experiment was to relate plasma CdMT to Cd-induced hepatic and renal functions. Reference values are given in Table 1, and all data were analyzed by the dose-response relationship to estimate the duration of Cd exposure necessary to induce hepatic and renal dysfunction as well as elevated plasma CdMT. Rabbits were given 130 g/day commercial pelleted food (CLEA CR-3; 0.06 mg Cd/g, 7 mg Cu/g, 31 mg Zn/g, 392 mg Fe/g, 20.3 mg Ca/g) and tap water ad libitum, and were injected s.c. on the back daily with cadmium chloride at a dose level of 0.3 mg Cd/kg, until they suffered from renal dysfunction. Blood and urine specimens were collected at 0, 4, 8, 10, 12, 13, 14 and 16 weeks of the experiments, by the time when animals suffered from renal dysfunction. From one rabbit which responded to Cd late, blood and urine specimens were collected, additionally every week until the 20th week of the experiments. One-day urine specimen was collected from each rabbit using a metabolic cage, avoiding fecal contamination. The plasma CdMT was used in the present study as a monitoring index for renal dysfunction instead of CdMT in proximal convoluted tubular lumen, because (1) it was technically difficult to measure fluid CdMT in the proximal convoluted tubular lumen in vivo because of such a small urine volume to be fractionated, and (2) the CdMT in the proximal convoluted tubular lumen right after passage through the glomeruli varies and is likely parallel with the plasma CdMT considering its easy passage through glomeruli because of its small molecular weight of 6000. We did not substitute CdMT in the urine for CdMT in renal tubular lumen because CdMT was reabsorbed at renal proximal tubular lumen with saturation process (Nomiyama and Foulkes, 1977). We also did not use total MT nor total Cd in plasma instead of plasma CdMT, because it is quite possible that the ratio of CdMT to total MT or the ratio of CdMT to total Cd all in the plasma varies with Cd-induced hepatic function and others.

Reference value 0 0 0 0 0 – – – 0 0 0 0 8 0 0 0 0 0

0 week (%) 0 0 0 0 0 0 – 0 – 31 0 0 0 0 0 0 0 15

5 weeks (%) 0 0 33 0 0 0 0 0 O 23 0 8 8 8 0 0 0 8

8 weeks (%) 17 11 10 23 8 8 23 0 0 25 0 20 18 10 0 9 11 -

10 weeks (%) 62 38 58 54 23 31 0 8 100 15 54 31 23 33 23 31 38 54       

 

    

   

 

           

12 weeks (%) 60 57 33 40 10 30 0 0 100 30 60 20 20 37 20 40 60 63

       

 

  

   

 

      

13 weeks (%)

90 80 50 90 60 40 11 10 90 20 90 30 0 44 44 100 90 80

               

   

   

                   

14 weeks (%)

a

–, represents no data available.  ,    and     represent a significantly elevated incidence of rabbits with a higher value at 5, 1 and 0.1% levels of probability, respectively.

Plasma cadmium 40 mg/l5 Plasma cadmium–metallothionein 20 mg/l5 Blood cadmium 1700 mg/l5 Plasma aspartate aminotransferase 50 IU/l5 Plasma alanine aminotransferase 70 IU/l5 Plasma urea nitrogen 20 mg/dl5 Plasma urea nitrogen/creatinine 15 5 Plasma uric acid 0.6 mg/dl5 Plasma creatinine 1.7 mg/dl5 Urine alkaline phosphatase 60 IU/g Cr5 Urine g-glutamyl transpeptidase 100 IU/g Cr5 Urine aspartate aminotransferase 20 IU/g Cr5 Urine alanine aminotransferase 15 IU/g Cr5 Urine N-acetyl-b-D-glucosaminidase 2 IU/g Cr5 Urine protein 1000 mg/g Cr 5 Urine amino acids 20 mmol/g Cr5 Urine glucose 200 mg/g Cr5 Incidence of urine blood (qualitative) +5

Parameters

Table 1 Dose-response relationship in 13 Male rabbits given subcutaneous injections of cadmium chloride at a dose level of 0.3 mg/kg daily over a period of 14 weeksa

K. Nomiyama et al. / Toxicology 129 (1998) 157–168 159

160

K. Nomiyama et al. / Toxicology 129 (1998) 157–168

The plasma CdMT was determined as follows: plasma was filtered through a Millex-HV13 filter unit (Millipore, a sterile 13 mm diameter filter with 0.45 mm pore size) (efficiency 98%), and then a 100 ml aliquot was separated to obtain the CdMT fraction by high performance liquid chromatography (HPLC) on a Superdex 75 high-performance gel filtration column using a high speed liquid chromatograph (Toyo Soda HLC-803A). The Cd in each fraction was determined with a flameless atomic absorption spectrophotometer (Hitachi Z-9000), and the CdMT in the plasma was estimated versus standard CdMT (Sigma M7641). It should be noticed also that almost all Cd is CdMT, when Cd of low molecular weight (6000–10 000) fraction of the plasma from rabbits exposed to Cd was determined by the flameless atomic absorption spectrophotometry. Plasma CdMT, which we determined by the above method, was the same as standard rabbit liver CdMT in the HPLC column profile (Fig. 1). We calculated the plasma CdMT by multiplying the plasma total Cd (mg/l) by the ratio of CdMT to total Cd, which was estimated from each HPLC column profile. Further, in the present paper, we employed mg Cd/l, as the units for CdMT, because we determined CdMT by a combination of HPLC and atomic absorption spectrophotometry,

Fig. 1. High performance liquid chromatography-atomic absorption spectrophotometry profiles. Solid and dotted lines represent for rabbit cadmium–thionein standard and plasma of a rabbit given daily subcutaneous cadmium at a dose level of 0.3 mg Cd/kg over a period of 16 weeks.

and, additionally, active Cd, but not protein MT, has been suggested to induce dysfunction of the renal tubules (Nordberg et al., 1985). If the need arises for readers to convert the mg Cd data into mg MT data, this can be done simply by multiplying an appropriate factor. However, we choose not to convert our data from mg Cd/l to mg MT/l, because of the difficulty in estimating what proportion of MT binds to Cd in our experiments. To determine blood Cd, blood was diluted first with distilled water five times, and then 1/200 volume of concentrated nitric acid was added. The solution was heated at 60–70°C for 5 min and centrifuged. Plasma was first diluted with 0.1 N nitric acid solution. Both blood and plasma specimens were analyzed for Cd with a combination of the standard addition method and flameless atomic absorption spectrophotometry with the use of simultaneous multi-element atomic absorption spectrophotometer Hitachi Z-9000. Plasma aspartate aminotransferase and alanine aminotransferase were determined with an automatic biochemical analyzer, Abbott VP, employing the GOT-FA test (Wako) and GPT-FA test (Wako). Plasma urea nitrogen was determined primarily to assess glomerular function but also to assess dysfunction in nitrogen metabolism in the liver, by employing a-Gent BUN (Abbott) and with an automatic biochemical analyzer (Abbott VP). Plasma uric acid was determined by employing a-Gent Uric Acid (Abbott) to assess dysfunction of nitrogen metabolism in the liver. Urinary enzymes such as aspartate aminotransferase and alanine aminotransferase were determined to assess the cellular injury of the proximal convoluted tubules, since these enzymes are located in the cytosol of the tubular cells (World Health Organization, 1991). Urine g-glutamyl transpeptidase and N-acetyl-b-D-glucosaminidase were assayed by employing a-gent GGT (Abbott) and the NAG Test (Shionogi) with an automatic biochemical analyzer (Abbott VP). The above urine enzymes are sensitive parameters to detect injury of brush border membrane and lysosomes of tubular cells. Urine protein and glucose were also analyzed by using Tonein TP1 (Otsuka) and the Glucose FA Test (Wako) with an automatic biochemical analyzer (Abbott VP). Urine amino

K. Nomiyama et al. / Toxicology 129 (1998) 157–168

acids and the plasma creatinine were estimated by the trinitrobenzene sulfonic acid (TNBS) method (Nomiyama, 1973) and the Folin-Wu method (Bonsnes and Taussky, 1945), respectively. The statistical significance of the Cd-induced toxic effects by weeks of Cd administration was analyzed by analysis of variance and the correlations between the two parameters were evaluated by Cricket Graph 1.3.2. The criteria for statistical significance was P 50.05, 0.01 or 0.001 (indicated by *, ** or *** in the figures).

3. Results

3.1. Temporal relationships between hepatic dysfunction, ele6ated plasma cadmium metallothionein and renal dysfunction in rabbits exposed to cadmium The plasma aspartate aminotransferase and alanine aminotransferase were markedly elevated after the 12th week (Fig. 2A and B, P B 0.01 and PB 0.05; Table 1, P B 0.01 except for alanine aminotransferase). Plasma urea nitrogen increased markedly after 12th week (Table 1, PB 0.05), but the ratio of plasma urea nitrogen to creatinine remained at the normal level during the entire experiment (Table 1). It should be noticed that there was a considerable variation in duration of Cd exposure necessary to induce hepatic dysfunction (10 – 20 weeks) as seen in Fig. 2A and B. The plasma CdMT increased after the 12th week of the experiments (Fig. 2C PB 0.01; Table 1, P B0.05 or 0.01). Plasma Cd and blood Cd increased markedly after the 12th week of the experiments (Table 1, P =0.01 and P B 0.001). In the 12th week, glucose and amino acids both in urine increased (Fig. 2D P B 0.001, Fig. 2E P B0.001, Table 1, P B0.05 – 0.001). Urine protein increased markedly also at 14th week of the experiment (Fig. 2F and Table 1, P B 0.05 for both). Plasma creatinine increased markedly after 12th week of the experiment (Fig. 2G, P B 0.001; Table 1, PB0.001). The urinary excre-

161

tions of g-glutamyl transpeptidase, an enzyme located at the brush border membrane (World Health Organization, 1991), and N-acetyl-b-Dglucosamini-dase, an enzyme located in the lysosomes of tubular cells (World Health Organization, 1991), also increased at the same week (Fig. 2H, PB 0.01; Fig. 2I P B0.05; Table 1, PB 0.01–0.001 for g-glutamyl transpeptidase and PB 0.05 for N-acetyl-b-D-glucosaminidase). The urinary excretion of aspartate aminotransferase, an enzyme located in the renal tubular cells (World Health Organization, 1991), also rose at 12 weeks (Fig. 2J, P B0.01; Table 1, PB 0.05 at 12 weeks only). Prevalence of hematuria was also elevated markedly after 12th week (Fig. 2K, PB0.01–0.001). The statistical analysis indicated that the Cdinduced hepatic and renal dysfunctions appeared mostly at 12 weeks of Cd exposure, together with plasma CdMT.

3.2. Relationships between plasma cadmium–metallothionein and toxic effects 3.2.1. Relationship between hepatic functions and plasma cadmium–metallothionein As shown in Fig. 3A, a very close correlation was noted between the plasma aspartate aminotransferase and the plasma CdMT (PB 0.001). The same was true for the plasma alanine aminotransferase and the plasma CdMT (Fig. 3B, PB 0.001). The plasma urea nitrogen and uric acid increased with the plasma CdMT (Fig. 3C and D, PB 0.001 for both). However, the ratio of the plasma urea nitrogen to creatinine remained at the same level regardless of the plasma CdMT (Fig. 3E). The above facts may indicate that Cd depresses the metabolism of nitrogen in the liver. 3.2.2. Relationship between plasma cadmium–metallothionein and renal functions The plasma creatinine was also closely correlated with the plasma CdMT (Fig. 3F, PB 0.001). As shown in Fig. 3G–I, protein, glucose and amino acids all in the urine were closely correlated with the plasma CdMT (PB 0.001 for

162

K. Nomiyama et al. / Toxicology 129 (1998) 157–168

Fig. 2. Plasma cadmium –metallothionein, hepatic and renal functions in 13 rabbits given cadmium at a dose level of 0.3 mg Cd/kg over a period of 20 weeks. Cadmium administration was terminated, right after hepatic and renal dysfunctions were detected.

K. Nomiyama et al. / Toxicology 129 (1998) 157–168

163

Fig. 3. Correlations between plasma cadmium–metallothionein and hepatic/renal function. *, ** and *** represent the statistical significance at 5, 1 and 0.1% levels, respectively.

164

K. Nomiyama et al. / Toxicology 129 (1998) 157–168

all). The same was true for some enzymes located at the brush border membrane (g-glutamyl trans-peptidase) (Fig. 3J, P B O.001), in the cytosol(aspartate aminotransferase and alanine aminotransferase) (Fig. 3K and L) and in the lysosomes (N-acetyl-b-D-glucosaminidase) (Fig. 3M, PB O.001).

4. Discussion

4.1. Temporal con6ergence of cadmium-induced hepatic and renal dysfunctions in animals chronically exposed to cadmium As seen in Fig. 2 and Table 1, hepatic dysfunction, elevated plasma CdMT level and renal dysfunction were all detected mostly at 12 weeks, in rabbits given cadmium chloride daily at a dose level of 0.3 mg Cd/kg. Nomiyama et al. (1979) reported previously that hepatic dysfunction (12th week) preceded renal dysfunction (15th week) in monkeys given oral Cd at a dose level of 300 mg Cd/g over a period of 1 year. Monkeys given oral Cd at a lower dose level (100 mg Cd/g) over a period of 9 years, also suffered from hepatic dysfunction (6 – 18th week) prior to renal dysfunction (48th – 91st week) (Nomiyama et al., 1987a). Nomiyama et al. (1973) also reported that rabbits given subcutaneous injections of Cd on the back daily at a dose level of 1.5, 6 or 15 mg Cd/kg, hepatic dysfunction (28th, 9th or 5th day for the 1.5, 6 or 15 mg Cd/kg group, respectively) were detected before or at the same time with renal dysfunction (28th, 12th or 9th day). Nomiyama and Nomiyama (1982) also found that hepatic dysfunction (9–12th week) and renal dysfunction (9 – 12th week) developed at the same time in rabbits given subcutaneous injections of Cd at a dose level of 0.5 mg Cd/kg over a period of 21 weeks. Dudley et al. (1985) indicated that hepatic injury was detected at the 6th week prior to renal toxicity at the 9th week in rats exposed to Cd. All the above data strongly support our hypothesis that hepatic dysfunction appears at the same time or a little earlier than renal dysfunc-

tion in animals exposed to Cd at different dose levels and in different animal species.

4.2. Mechanism of ele6ated plasma cadmium–metallothionein in animals exposed to cadmium The plasma CdMT in animals exposed to Cd increased at the same time as the Cd-induced hepatic dysfunction occurred (Fig. 2, Table 1), and the increase in the plasma CdMT was proportional to the degree of the Cd-induced hepatic dysfunction (Fig. 3). These findings suggest that hepatic CdMT is transferred to the blood stream upon the Cd-induced hepatic dysfunction. Tanaka (1982) reported that rats given Cd suffered from renal dysfunction after massive CdMT release from the liver into blood upon carbon tetrachloride administration. Nomiyama and Nomiyama (1994) later found that the degree of renal dysfunction depended on the plasma CdMT level, which varied both by accumulated CdMT levels in the liver and the severity of carbon tetrachloride-induced hepatic dysfunction. Dudley et al. (1985) suggested that the Cd-induced hepatic injury might result in hepatic release of CdMT, with its subsequent translocation to the kidneys to produce renal injury. Chan et al. (1992) also suggested that hepatic MT, but not CdMT, was the source of the plasma MT. It was also the same for humans: Shiraishi et al. (1985) found that the urine Cd, which might be closely related to CdMT in renal tubular lumen, was elevated simultaneously with high plasma aspartate aminotransferase in suspected Itai-Itai disease patients, who lived in a Cd-polluted area in Japan. The above reports support our finding that hepatic dysfunction is associated with elevated plasma CdMT. However, hepatic dysfunction is not always necessary for plasma CdMT elevation. Cadmium is known to induce apoptosis in hepatic cells (Lyman and Beyersmann, 1994). Cd-induced apoptosis of hepatic cells may also accelerate release of CdMT from the liver into blood stream so as to induce in animals or humans renal dysfuction.

K. Nomiyama et al. / Toxicology 129 (1998) 157–168

4.3. Mechanism of cadmium– metallothionein-induced renal dysfunction The present study indicated that the Cd-induced renal dysfunction developed when the plasma CdMT rose markedly (Fig. 2). The degree of renal dysfunction was proportional to the level of the plasma CdMT (Fig. 3). This suggests that CdMT in the plasma passes through the glomeruli into the renal tubular lumen and injures brush border membrane of the proximal convoluted tubules and that the degree of renal dysfunction is proportional to the CdMT concentration in the tubular lumen. It has been reported that a single dose of CdMT can induce renal dysfunction: Cherian and Shaikh indicated as early as 1974 that CdMT, injected intravenously into rats, was transported mainly to the kidneys and then excreted into the urine in the chemical form of CdMT. Tanaka et al. (1975) also reported similar findings. Later, Nomiyama and Foulkes (1977) examined the kinetics of renal tubular reabsorption of CdMT. Nordberg (1972), Nordberg et al. (1985) clearly stated that in mice CdMT was more toxic to the kidneys than Cd itself. Webb (1975) suggested that the renal toxicity of Cd could be caused by free Cd liberated from CdMT. Cherian et al. (1976) reported that in rats necrosis of renal proximal tubular lining cells was observed within 24 h following intraperitoneal administration of CdMT. Cherian and Goyer (1978) proposed two hypotheses for the mechanism of the Cd-induced renal dysfunction, (1) CdMT within the renal tubular lining cells becomes saturated, and the renal cell injury results from unbound Cd; and (2) CdMT somehow becomes extracellular and exerts a direct toxic effect on the cell membranes. Even though several reports have subsequently indicated that single or repeated doses of CdMT can induce renal dysfunction in animals, they did not deal with the dose-effect nor dose-response relationship between an elevated plasma (or tubular) CdMT (but not total MT) and the Cd-induced renal dysfunction, in animals exposed to Cd. Independent of Cherian et al. (1976), Shaikh and Hirayama (1979) reported that total MT, but not CdMT, was markedly increased in the plasma at

165

the 10th week of Cd exposure, and that glucosuria and proteinuria then became severe during the next 4 weeks, in rats given subcutaneous Cd. Squibb et al. (1979) gave CdMT intravenously to rats and found CdMT in acidic lysosome of proximal tubular cells, where CdMT may be broken down into toxic ionic Cd. On the other hand, Suzuki (1980) reported also that the content of Cd in the urine, which might be relatively in proportional to CdMT in renal tubules, increased probably upon the Cd-induced hepatic dysfunction, and was then followed by renal dysfunction, in rats given Cd daily over a period of 25 weeks. Later, Nordberg et al. (1985) presented the following model for Cd-induced renal dysfunction: CdMT, transported from the liver to the kidney, was broken down into Cd and amino acids, but that liberated Cd ion, was found to be detoxified by binding with MT produced in renal cells. Therefore, it has been believed that renal dysfunction did not develop because renal cells could produce enough MT to detoxify the Cd. On the other hand, Dorian et al. (1992) described an interesting alternative: CdMT-inducedcellular effects began with the formation of pinocytolic vesicles of the luminal surface followed by focal mitochondrial changes. Further, Sudo et al. (1996) clearly demonstrated a direct effect of CdMT on brush border membrane by a close association between cellular membranebinding Cd in the kidneys and the decreases in renal leucine aminopeptidase, which are located in brush border membrane of tubular cells (Sudo and Tanabe, 1984). Chan et al. (1993) transplanted livers from Cdinjected rats to age-matched control healthy rats, and found that renal Cd and MT levels were markedly increased, about 100 mg/l CdMT was detected in the plasma of the recipient rats. There was an increase in blood urea nitrogen levels in the rats transplanted with Cd containing livers, and both necrosis and inflammation were observed in the epithelial cells of the proximal tubules within the kidney, as occurs typically in chronic Cd toxicity. They suggested that the major source of renal Cd in chronic Cd exposure could be derived from hepatic Cd which is transported in the form of CdMT in the blood plasma.

166

K. Nomiyama et al. / Toxicology 129 (1998) 157–168

Fig. 4. Proposed model of Cd-induced renal dysfunction (Nomiyama and Nomiyama, 1998). (1) Cd level in the renal cells is not responsible for Cd-induced membrane dysfunction of the proximal tubular cells. (2) CdMT is taken up at brush border membrane of proximal tubular cells and may be broken down into ionic Cd and amino acids in acidic lysosome, and then ionic Cd directly injures brush border membrane, but independent of total Cd or renal ability to produce MT to detoxify Cd in the renal cortex. (3) The critical concentration of plasma CdMT, which may be proportional to the CdMT level in tubular lumen, to induce renal dysfunction is 80 mg Cd/1 in rabbits.

Considering our data and the reports of Dorian et al. (1992), Chan et al. (1993), Sudo et al. (1996), our proposal on the mechanism of the Cd-induced renal dysfunction (Fig. 4) differs in some parts from Nordberg et al. (1985): CdMT is taken up at brush border membrane of proximal tubular cells and may be broken down into ionic Cd and amino acids in acidic lysosome (Squibb et al., 1979) relatively close to the brush border membrane of proximal tubular cells. Then ionic Cd, the concentration of which may be proportional to CdMT in the tubular lumen, directly injures brush border membrane (Sudo et al. 1996), independently of the

total Cd or the renal ability to produce MT to detoxify Cd in the renal cells. Thus, the Cd-induced renal dysfunction could recover upon the decrease in plasma CdMT by diminishing CdMT release from the liver by treating the Cd induced hepatic dysfunction, independent of Cd concentration in the renal cortex (Nomiyama et al., 1973). It should be noted also here that decreased Cd toxicity by pretreatment of Cd in animals may be bought by the decreased reabsorption of CdMT at renal tubules due to saturation of epithelial binding site of renal tubular cells for CdMT uptake (Vestergaad and Shaikh, 1994).

K. Nomiyama et al. / Toxicology 129 (1998) 157–168

4.4. Critical concentration of plasma cadmium–metallothionein for inducing renal dysfunction The critical concentrations of the plasma CdMT to induce proteinuria (urine protein above 1500 mg/g Cr), glucosuria (urine glucose above 2500 mg/g Cr) and aminoaciduria (urine amino acids above 70 mmol/g Cr) were 80, 75 and 85 mg Cd/l, respectively, as shown in Fig. 3G – I. These values imply that the critical concentration of the plasma CdMT to induce renal dysfunction is 80 mg Cd/l. Nomiyama and Nomiyama (1994) also estimated the critical concentration of the plasma CdMT as 50 mg Cd/l in rabbits given Cd at a dose level of 0.5 mg Cd/kg daily for 25 weeks. It needs further studies to settle the critical concentration of the plasma CdMT in animals and humans exposed to Cd. Acknowledgements The authors appreciate greatly the help of Dr E.C. Foulkes and Mr A.J. Smith for their editing this paper, and of Mr T. Tsuchiya for his carefully maintaining the animals. The present study was partly supported by research grants from the Japanese Ministry of Education, Science and Culture (09470103) and the Japan Environment Agency in 1997. References Bonsnes, R.W., Taussky, H.H., 1945. On the colorimetric determination of creatinine by the Jaffe reaction. J. Biol. Chem. 158, 581 – 591. Chan, H.M., Stash, M., Zalups, R.K., Cherian, M.G., 1992. Exogenous metallothionein and renal toxicity of cadmium and mercury in rats. Toxicology 76, 15–26. Chan, H.M., Zhu, L.F., Zhong, R., Grant, D., Goyer, R.A., Cherian, M.G., 1993. Nephrotoxicity in rats following liver transplantation from cadmium-exposed rats. Toxicol. Appl. Pharmacol. 123, 89–96. Cherian, M.G., Goyer, R.A., 1978. Role of metallothionein in disease. Ann. Clin. Lab. Sci. 8, 91–94. Cherian, M.G., Goyer, R.A., Delaquerrier-Richardson, L., 1976. Cadmiummetallothionein-induced nephropathy. Toxicol. Appl. Pharmacol. 38, 399–408.

167

Dorian, C., Gattone II, V.H., Klaassen, C.D., 1992. Renal cadmium deposition and injury as a result of accumulation of cadmium – metallothionein (CdMT) by the proximal convoluted tubules — a light microscopic autoradiography study with 109CdMT. Toxicol. Appl. Pharmacol. 114, 173 – 181. Dudley, R.E., Gammal, L.M., Klaassen, C.D., 1985. Cadmium-induced hepatic and renal injury in chronically exposed rats: likely role of hepatic cadmiummetallothionein in nephrotoxicity. Toxicol. Appl. Pharmacol. 77, 414 – 427. Lyman, R.D., Beyersmann, D., 1994. Effects of zinc and cadmium on apoptotic DNA fragmentation in isolated bovine liver nuclei. Environ. Health Perspect. 102 (Suppl 3), 269 – 271. Nomiyama, K., 1973. Early detection of chronic cadmium poisoning. Jpn. J. Med. 2587, 132. Nomiyama, K., Foulkes, E.C., 1977. Reabsorption of filtered cadmium – thionein in the rabbit kidney. Proc. Soc. Exp. Biol. Med. 156, 97 – 99. Nomiyama, K., Nomiyama, H., 1982. Tissue metallothionein in rabbits chronically exposed to cadmium, with special reference to the critical concentration of cadmium in the renal cortex. In: Foulkes, E.C. (Ed.), Biological Role of Metallothionein. Elsevier, New York, pp. 47 – 67. Nomiyama, K., Nomiyama, H., 1986. Critical concentration of ‘unbound’ cadmium in the rabbit renal cortex. Experienta 42, 149. Nomiyama, K., Nomiyama, H., 1994. Cadmium-induced renal dysfunction. Dtsch. Med. Wschr. (Jpn.) 16, 189 – 201. Nomiyama, K., Nomiyama, H., 1998. Cadmium-induced renal dysfunction: new mechanism, treatment and prevention. J. Trace Elem. Exp. Med. 11, 275 – 288. Nomiyama, K., Sato, C., Yamamoto, A., 1973. Early signs of cadmium intoxication in rabbits. Toxicol. Appl. Pharmacol. 24, 625 – 635. Nomiyama, K., Nomiyama, H., Nomura, Y., Taguchi, T., Matsui, K., Yotoriyama, M., et al., 1979. Effects of dietary cadmium on rhesus monkeys. Environ. Health Perspect. 28, 223 – 243. Nomiyama, K., Nomiyama, H., Akahori, F., Masaoka, T., 1987a. Experimental studies on cadmium health effects in monkeys exposed for 9 years. (6) Hepatic effects. Kankyo Hoken Rep 53, 44 – 55. Nomiyama, K., Akahori, E., Nomiyama, H., Nasaoka, T., Arai, S., Nomura, Y., et al., 1987b. Dose – effect and dose – response relationships of a years dietary cadmium on rhesus monkeys. Kankyo Koken Report 53, 1 – 86; 13 – 134. Nordberg, G.F., 1972. Cadmium metabolism and toxicity: experimental studies on mice. Environ. Physiol. Biochem. 2, 7 – 36. Nordberg, G.F., Kjellstrom, T., Nordberg, M., 1985. Kinetics and metabolism. In: Friberg, L., Elinder, C.-G., Kjellstrom, T., Nordberg, G.F. (Eds.), Cadmium and Health: A Toxicological and Epidemiological Appraisal. Exposure, Dose and Mechanism, vol. 1. CRC Press, Boca, Raton, Florida, p. 142.

168

K. Nomiyama et al. / Toxicology 129 (1998) 157–168

Shaikh, Z.A., Hirayama, K., 1979. Metallothionein in the extracellular fluids as an index of cadmium toxicity. Environ. Health Perspect. 28, 267–271. Shiraishi, K., Niimura, T., Uetake, H., 1985. Blood and urine tests of suspected Itai–Itai disease patients. (1) Cadmium in urine. Kankyo Hoken Rep 51, 147–149. Squibb, K.S., Ridlington, J.W., Carmichael, N.G., Fowler, B.A., 1979. Early cellular effects of circulating cadmium– thionein on kidney proximal tubules. Environ. Health Perspect. 28, 287 – 296. Sudo, J., Tanabe, T., 1984. Distribution of aminopeptidases in various nephron segments isolated from rat kidney. Chem. Pharm. Bull. 32, 3235–3243. Sudo, J., Hayashi, T., Kimura, S., Kakuno, K., Terui, J., Takashima, K, Soyama, M., 1996. Mechanism of nephrotoxicity induced by repeated administration of cadmium chloride in rats. J. Toxicol. Environ. Health 48, 333–348. Suzuki, Y., 1980. Cadmium metabolism and toxicity in rats

.

after long-term subcutaneous administration. J. Toxicol. Environ. Health 6, 469 – 482. Tanaka, K., Sueda, K., Onosaka, S., Okahara, K., 1975. Fate of 109Cd-labeled metallothionein in rats. Toxicol. Appl. Pharmacol. 33, 258 – 266. Tanaka, K., 1982. Effects of hepatic disorder on the fate of cadmium in rats. In: Foulkes, E.C. (Ed.), Biological Role of Metallothionein. Elsevier, New York, pp. 237 – 249. Vestergaad, O., Shaikh, Z.A., 1994. The nephrotoxicity of intravenously administered cadmium – thionein: effect of dose, mode of administration, and preexisting renal cadmium burden. Toxicol. Appl. Pharmacol. 126, 240 – 247. Webb, M., 1975. Toxicity of cadmium – thionein. In: Proceedings of the International Conference on Heavy Metals in the Environment, Toronto, B-47-49. World Health Organization, 1991. Principles and methods for the assessment of nephrotoxicity associated with exposure to chemicals. World Health Organization, Geneva, p. 178.