- Email: [email protected]
Attenuation of Cadmium-Induced Liver Injury in Senescent Male Fischer 344 Rats: Role of Kupffer Cells and Inflammatory Cytokines
Toxicology and Applied Pharmacology 162, 68 –75 (2000) doi:10.1006/taap.1999.8833, available online at http://www.idealibrary.com on
Attenuation of Cadmium-Induced Liver Injury in Senescent Male Fischer 344 Rats: Role of Kupffer Cells and Inflammatory Cytokines Tetsuo Yamano,* Lacinda A. DeCicco,† and Lora E. Rikans† *Osaka City Institute of Public Health and Environmental Sciences, Osaka, Japan; and †Department of Pharmaceutical Sciences, University of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma 26901 Received June 28, 1999; accepted October 21, 1999
information relevant to metal toxicity as well as to the development of liver damage from other toxicants. Thus, Cd has become an important experimental hepatotoxicant. In a previous study we found that senescent rats are much more resistant to Cd-induced acute hepatotoxicity than are their younger counterparts and that age-associated changes in hepatic metallothionein and glutathione do not explain this phenomenon. These results are consistent with previous findings in which age-associated differences in these protective factors did not account for increased liver damage from Cd in 6-month-old rats compared with 1- or 2-month-old rats (Yamano et al., 1998). Instead, the results suggested that the age-dependent shift in sensitivity was due to changes in Kupffer cell and/or neutrophil activity. Thus, the current study was undertaken to examine the role of inflammatory processes in the postmaturational effects of aging on Cd hepatotoxicity. The involvement of Kupffer cells in Cd-mediated liver damage is suggested by the results of several investigations. Kupffer cell activation, demonstrated as increased cytoplasmic vacuolization or increased carbon clearance, has been observed after Cd administration to rats (Hoffman et al., 1975; Sauer et al., 1997). In addition, the administration of gadolinium chloride, an inhibitor of Kupffer cell function, alleviates acute Cd hepatotoxicity in vivo without substantially affecting hepatic Cd accumulation or hepatic metallothionein and glutathione contents (Sauer et al., 1997; Yamano et al., 1998). Inhibition by gadolinium was not observed in isolated hepatocytes, indicating that the prevention of Cd hepatotoxicity in vivo depended on the participation of nonparenchymal cells, i.e., Kupffer cells (Badger et al., 1997). Although the mechanisms by which Kupffer cells contribute to Cd-induced hepatotoxicity are not entirely clear, it is known that activated Kupffer cells release a number of inflammatory mediators (e.g., cytokines, chemokines, adhesion molecules) that initiate a cascade of cellular and humoral responses leading to inflammation and secondary damage to the liver. Infiltration of neutrophils at the site of injury is a common observation in Cd-induced hepatic necrosis (Stowe et al., 1972). The migration and activation of neutrophils are regulated by cytokines, chemokines, and adhesion molecules (Bag-
Attenuation of Cadmium-Induced Liver Injury in Senescent Male Fischer 344 Rats: Role of Kupffer Cells and Inflammatory Cytokines. Yamano, T., DeCicco, L. A., and Rikans, L. E. (2000). Toxicol. Appl. Pharmacol. 162, 68 –75. In the previous study we showed that senescent male Fischer 344 rats were resistant to Cd-induced hepatotoxicity compared with young-adult rats. In the present study we investigated the role of Kupffer cells and inflammatory cytokines in this effect of aging. The phagocytic activity of Kupffer cells, determined as the removal of carbon from blood, was stimulated by the administration of a hepatotoxic dose of Cd (3 mg/kg sc) in young-adult (5 months) rats but not in old (28 months) rats. Hepatic concentrations of interleukin (IL)-1 and cytokine-induced neutrophil chemoattractant (CINC), but not of tumor necrosis factor-␣ or IL-6, were elevated in young rats treated with Cd. In old rats, however, the increase in IL-1 produced by Cd was not statistically significant and the increase in CINC was much lower than in youngadult rats. Pretreatment with gadolinium chloride or cyclosporin A inhibited the elevations in hepatic cytokines and attenuated Cd-induced liver damage, assessed on the basis of serum alanine aminotransferase and sorbitol dehydrogenase activities. Cd-induced hepatotoxicity in the different treatment groups correlated well with hepatic levels of CINC (r ⴝ 0.98, p < 0.001) but not with those of IL-1. The results suggest that (1) Kupffer cell activation is essential for inflammatory liver damage from Cd, (2) IL-1 and CINC are important mediators of the inflammatory response induced by Cd, and (3) the attenuation of Cd-induced liver injury in senescent rats is caused by an impairment in Kupffer cell activation, leading to a lower production of CINC and less inflammatory liver injury. © 2000 Academic Press Key Words: Cadmium; cyclosporin A; gadolinium chloride; cytokine-induced neutrophil chemoattractant; interleukin IL-1; inflammatory liver injury; Kupffer cells.
Toxicity from chronic exposure to low doses of Cd is manifested mainly as chronic renal tubular disease and chronic pulmonary disease. On the other hand, acute exposure to Cd results primarily in liver damage, because the metal initially accumulates in the liver. Studies of Cd-induced hepatotoxicity in laboratory animals have contributed valuable mechanistic 0041-008X/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
68
Cd HEPATOTOXICITY IN OLD AGE: ROLE OF IMMUNE RESPONSE
giolini et al., 1994; Jaeschke and Smith, 1997; Ohkubo et al., 1998). Increased expression of several of these mediators, including tumor necrosis factor (TNF)-␣, interleukin (IL)-1␣, IL-1, IL-6, interferon-␥, intercellular adhesion molecule-1, macrophage inflammatory protein-2, and cytokine-induced neutrophil chemoattractant (CINC), has been demonstrated in rodent liver or hepatocytes exposed to Cd (Kayama et al., 1995; Dong et al., 1998). Moreover, pretreatment of mice with anti-TNF-␣ antibodies prevented Cd-induced secretion of acute phase proteins and focal inflammation in the liver (Kayama et al., 1995). These results indicate that inflammatory cytokines are expressed following Cd exposure and are responsible for the manifestations of inflammation observed in Cdinduced hepatotoxicity. In a previous study, we showed that the difference between young and mature rats in sensitivity to Cd-induced liver damage was decreased substantially by the inactivation of Kupffer cells or the depletion of neutrophils (Yamano et al., 1998). However, little is known about the effects of postmaturational aging on hepatotoxicant-induced inflammatory responses, such as Kupffer cell activation, cytokine production, or migration of neutrophils to the site of injury. In this study we compared rates of carbon clearance, a measure of Kupffer cell activity, and hepatic levels of cytokines in young-adult and old rats treated with Cd. METHODS Animals. Male Fischer 344 (F344) rats were obtained from the colony supported by the National Institute on Aging at Harlan Sprague Dawley Inc. (Indianapolis, IN). Young-adult rats were 5 months and old rats were 25 to 28 months of age. Initial experiments to determine the time course of cytokine elevations in liver and experiments with inhibitors were performed using younger rats (F344 males about 3 months old). The rats were acclimated to our facilities for 1 to 2 weeks before use, and they were fed and housed as described previously (Rikans et al., 1996). Treatments. Rats were administered saline or CdCl 2 (3 mg Cd/kg body wt) dissolved in saline as sc injections (2.5 ml/kg). Food was withdrawn after treatment. In some studies the rats were pretreated with either cyclosporin A (Sigma Chemical Co., St. Louis, MO) or gadolinium chloride. Cyclosporin A (5 mg/kg) was injected intramuscularly 24 and 3 h before treatment with Cd or saline (Matsuda et al., 1998), and gadolinium chloride (10 mg/kg) was injected iv 24 h before treatment (Yamano et al., 1998). The rats were killed under pentobarbital anesthesia (30 mg/kg ip) for collection of blood and liver samples. The livers were frozen immediately in liquid nitrogen and kept at ⫺70°C for subsequent analysis. Colloidal carbon clearance. The clearance of colloidal carbon from the blood was determined as described by Szabo (1983) and Sauer et al. (1997) with slight modification. The carbon solution was prepared as follows: Commercial ink (Pelikan drawing ink A) was dialyzed 48 h against distilled water through a semipermeable membrane (12,000 MW cut-off). The dialysate was centrifuged at 5000 rpm and the carbon content in the supernatant was determined by weight. The supernatant was diluted with distilled water and NaCl was added to make final concentrations equal to 50 mg/ml of carbon and 0.85% NaCl. Rats were administered CdCl 2 or saline with or without gadolinium chloride pretreatment, as described above. Carbon solution (160 mg/kg) was injected into the tail vein under pentobarbital anesthesia (50 mg/kg ip) at 0.5, 1.5, 3, or 6 h after treatment with Cd or saline. Blood samples were
69
collected from the other side of the tail vein just before carbon injection and at 3, 6, 12, and 24 min after carbon injection. Each sample (44.7 l, collected in heparinized capillary tubes) was added to 2 ml of a 0.1% Na 2CO 3 solution and the absorbance of the hemolyzed blood was determined at 640 nm against a blank containing blood drawn prior to carbon injection. Carbon concentrations in blood were calculated from a standard curve of known carbon concentrations. Serum activities of hepatic enzymes. Serum alanine aminotransferase (ALT) and sorbitol dehydrogenase (SDH) activities were determined spectrophotometrically, using diagnostic kits from Sigma Chemical Co. Cytokine assays. Portions of liver were homogenized in nine parts of phosphate-buffered saline (PBS), pH 7.4, containing 2 mM of phenylmethylsulfonyl fluoride and 1 g/ml each of antipain, leupeptin, and pepstain A (all from Sigma Chemical Co.). The homogenates were centrifuged for 15 min at 15,000g at 8°C, and the supernatants were removed and stored at ⫺70°C until the assays were performed. Cytokine concentrations were determined using enzyme-linked immunosorbent assay (ELISA) procedures. TNF-␣ and IL-6 were measured using Cytoscreen ELISA kits from Biosource International (Camarillo, CA), and IL-1 was determined by a rat ELISA kit from Endogen (Woburn, MA). CINC concentrations also were quantified by ELISA. Microtiter plates were coated overnight at 4°C with polyclonal rabbit anti-rat CINC antibody (Pepro Tech Inc., Rocky Hill, NJ), using 100 l/well of a 1000 ng/ml solution in carbonate buffer, pH 9.6. After washing three times with PBS containing 0.05% Tween-20, blocking solution (PBS containing 1% BSA, 300 l/well) was added. The plates were sealed, incubated 1 h at 37°C, and washed. Recombinant rat CINC (Pepro Tech Inc.) or samples to be analyzed, diluted appropriately with PBS containing 0.1% BSA and 0.05% Tween-20, were added (100 l/well). The plates were sealed and incubated for 1 h at room temperature. After washing, biotinylated polyclonal goat anti-rat CINC antibody (R & D systems, Minneapolis, MN) in PBS containing 0.1% BSA and 0.05% Tween-20 was added (100 ng/ml, 100 l/well), and the plates were sealed and incubated for 1 h at room temperature. Following another washing, horseradish peroxidase-conjugated streptavidin (Pierce, Rockford, IL) was added (100 ng/ml, 100 l/well), and the plates were sealed and incubated for 20 min at room temperature. The plates were washed and incubated for 15 to 30 min in the dark with 100 l/well of o-phenylenediamine dihydrochloride buffer solution for ELISA (Sigma Chemical Co.). The enzymatic reaction was stopped by the addition of 50 l of 4 N HCl and the plates were read at 490 nm. The concentrations of CINC in liver fractions were calculated from a standard curve. The assay was linear for CINC concentrations between 7.8 and 500 pg/ml (0.4 to 25 ng/g liver), and the recovery of added recombinant rat CINC from liver fractions ranged from 89 to 112%. Statistics. Results are expressed as means ⫾ SE. Data were analyzed by Student’s t test, the Mann–Whitney rank sum test, one-way analysis of variance followed by Tukey’s test, or the Kruskal–Wallis test as appropriate.
RESULTS
The influence of Cd on the phagocytic activity of Kupffer cells in vivo was assessed by determining the removal of colloidal carbon from the blood. Carbon clearance in Cdtreated vs. control rats was significantly increased 1.5 h after Cd administration, and the clearance became progressively greater after 3 and 6 h. Because Cd-induced liver damage, indicated by elevated serum SDH levels, was evident by 6 h, all subsequent measurements of carbon clearance were determined 3 h after Cd administration. Figure 1 shows carbon concentrations in blood at 3, 6, 12, and 24 min after carbon injection. The value shown at 0 min is a theoretical initial concentration calculated by dividing the dose of carbon administered (160 mg/kg body wt) by the blood volume (67 ml/kg
70
YAMANO, DECICCO, AND RIKANS
FIG. 1. Effects of Cd and gadolinium chloride on carbon clearance. Young male F344 rats were treated with saline (Cont) or CdCl 2 (Cd; 3 mg/kg sc) 3 h prior to colloidal carbon injection (160 mg/kg iv), and blood carbon concentrations were determined at 3, 6, 12, and 24 min. Some rats were pretreated with gadolinium chloride (Gd; 10 mg/kg iv) 24 h prior to treatment with saline or Cd. Dotted lines denote estimated initial rates of clearance using a calculated value (2.4 mg/ml) for the initial carbon concentration. Values are means ⫾ SE for four to six rats. There are statistically significant differences in 3-min carbon concentrations between Cont vs. Cd ( p ⬍ 0.01), Cont vs Gd ( p ⬍ 0.05), and Cd vs Gd ⫹ Cd ( p ⬍ 0.001).
body wt) (Creskoff et al., 1949). There appeared to be two phases of carbon removal, an initial phase with a fast rate and a second phase with a slower rate. Initial clearance rates could not be calculated accurately because blood samples taken earlier than 3 min yielded erratic values for carbon concentration; however, Cd treatment caused a substantial decrease in 3-min carbon concentrations (0.84 ⫾ 0.09 vs. 1.42 ⫾ 0.05 in Cdtreated rats vs. saline-treated control rats, p ⬍ 0.01), suggesting that Cd treatment increased the initial rate of clearance. In contrast, pretreatment with gadolinium chloride increased the concentration of carbon remaining at 3 min (1.93 ⫾ 0.16 vs. 1.42 ⫾ 0.05 mg/ml in rats gadolinium-pretreated rats vs. control rats, p ⬍ 0.05, and 1.69 ⫾ 0.10 vs 0.84 ⫾ 0.09 mg/ml in rats receiving gadolinium ⫹ Cd vs. rats receiving Cd only, p ⬍ 0.001), suggesting a decrease in the initial clearance. The slow phase of carbon removal also was affected by Cd administration, but to a lesser extent. The effect of animal age on Kupffer cell activation was determined by comparing the effect of Cd on carbon removal in young-adult and old rats (5 and 28 months of age, respectively). The dose of Cd (3.0 mg/kg) used in these experiments was the same as the one used in the study demonstrating a marked attenuation of Cd-induced liver damage in old rats compared with young-adult rats. The amount of carbon remaining in the blood 3 min after carbon injection was similar in young-adult rats and old rats (1.89 ⫾ 0.15 and 1.67 ⫾ 0.17 mg/ml), suggesting that the initial phase of carbon clearance in control rats was unaffected by aging (Fig. 2). In contrast, the 3-min carbon levels were much lower in young-adult rats treated with Cd than in old rats treated with Cd (0.62 ⫾ 0.04 vs 1.56 ⫾ 0.14 mg/ml, p ⬍ 0.05), indicating that the stimula-
tion by Cd of carbon removal was more pronounced in youngadult rats than in old rats. In fact, no effect of Cd on carbon clearance was observed in the old animals. Concentrations of TNF-␣, IL-1, IL-6, and CINC were determined in liver and serum of young rats at various times after treatment with Cd or saline. Cd administration caused significant increases in hepatic levels of IL-1 and CINC (Fig. 3). A twofold elevation in IL-1 was apparent 6 h after Cd treatment. IL-1 was still elevated at the 12-h time point but returned to normal by 24 h. The rise in hepatic CINC concentration followed the rise in IL-1, beginning 12 h after Cd treatment and continuing for at least 24 h. Hepatic CINC levels at 24 h were three times greater in Cd-treated rats than in control rats. Serum SDH and ALT activities are also shown in Fig. 3, so that time courses for the increases in these cytokine levels can be compared with the development of hepatotoxic-
FIG. 2. Effect of Cd on carbon clearance in young-adult (5 months) and old (28 months) male F344 rats. The rats were treated with saline (Cont) or CdCl 2 (Cd; 3 mg Cd/kg sc) 3 h prior to colloidal carbon injection (160 mg/kg iv), and blood carbon concentrations were determined at 3, 6, 12, and 24 min. Dotted lines denote estimated initial rates of clearance using a calculated value (2.4 mg/ml) for the initial carbon concentration. Values are means ⫾ SE for four to six rats. There are statistically significant differences in 3-min carbon concentrations between young-adult Cont vs young-adult Cd ( p ⬍ 0.05) and young-adult Cd vs old Cd ( p ⬍ 0.05).
71
Cd HEPATOTOXICITY IN OLD AGE: ROLE OF IMMUNE RESPONSE
FIG. 3. Cd-mediated increases in hepatic IL-1 and CINC concentrations and in serum ALT and SDH activities. Young male F344 rats were treated with saline (Cont) or CdCl 2 (Cd; 3 mg/kg sc). IL-1 and CINC concentrations were measured by ELISA procedures, and serum activities of liver enzymes were measured spectrophotometrically. Values are means ⫾ SE for five to seven rats. *Effect of Cd treatment is significant, p ⬍ 0.05. **Effect of Cd treatment is significant, p ⬍ 0.01.
ity. Cd treatment did not affect hepatic concentrations of TNF-␣ or IL-6 (data not shown). The levels of TNF-␣, IL-1, and IL-6 in serum of control and Cd-treated rats were close to the limits of detection an did not correlate, either positively or negatively, with the levels in liver. In contrast, CINC levels in serum were easily detectable, and increases produced by Cd treatment paralleled those seen in liver (data not shown). Since hepatic concentrations of both IL-1 and CINC were elevated 12 h after Cd treatment, the effects of aging on Cd-mediated increases in the levels of these cytokines were determined at the 12-h time point. As expected, Il-1 levels were significantly increased in young-adult rats as a consequence of Cd treatment; however, the effect of Cd on IL-1 was not significant in old rats (Table 1). Likewise, hepatic CINC concentrations were 4.5-fold greater in Cd-treated young-adult rats than in control rats of the same age, whereas CINC levels in old rats were not increased significantly by Cd treatment. Serum SDH activities at 12 h were higher in youngadult rats than in old rats, confirming our earlier results that Cd-induced hepatotoxicity is significantly diminished in old age.
In order to explore the relationships between Cd-induced hepatotoxicity and Cd-mediated increases in hepatic IL-1 and CINC levels, rats were treated with gadolinium chloride, a Kupffer cell inhibitor, or cyclosporin A, an immunosuppressant, prior to Cd administration. Pretreatment with the inhibitors almost completely abrogated the hepatotoxicity, as shown by marked suppression of serum ALT and SDH activities 24 h after Cd treatment (Fig. 4). In addition, the inhibitors diminished the Cd-mediated elevations in hepatic IL-1 and CINC concentrations. The 12-h values for IL-1 and CINC are shown in Table 2. Gadolinium chloride completely suppressed the increase in IL-1 and partially suppressed the increase in CINC. Although cyclosporin A also suppressed IL-1 completely, its effect on CINC was not statistically significant at 12 h. The depression by inhibitors of hepatic CINC levels in Cd-treated rats was more pronounced at 24 h, and both inhibitors produced significant decreases in CINC at this time point (not shown). The 12-h concentrations of hepatic CINC in the various groups of rats correlated well with Cd-induced hepatotoxicity, measured as the leakage of SDH into the bloodstream (Fig. 5); however, correlations of IL-1 with hepatotoxicity were not significant. The severity of liver damage produced by Cd in young (3 months), young-adult (5 months), and old (25 to 28 months) rats also correlated well with hepatic CINC but not IL-1 levels (not shown). DISCUSSION
The mechanisms of acute Cd-mediated hepatotoxicity involves two pathways, one for the initial injury produced by direct effects of Cd and the other for the subsequent injury produced by the inflammatory response. Primary injury probably results from the binding of Cd to sulfhydryl groups on critical molecules in mitochondria, causing oxidative stress and the mitochondrial membrane permeability transition (Mu¨ller, 1986; Koizumi et al., 1994; Strubelt et al., 1996). Secondary injury from acute Cd exposure is thought to occur from the activation of Kupffer cells and a complex series of interactive TABLE 1 Effect of Age on Cd-Induced Elevations of Hepatic IL-1, Hepatic CINC, and Serum SDH Age (months)
Treatment
Hepatic IL-1 (ng/g)
Hepatic CINC (ng/g)
Serum SDH (IU/L)
5 5 25 25
Control Cd Conrol Cd
1.32 ⫾ 0.14 2.17 ⫾ 0.25 a 1.77 ⫾ 0.19 2.71 ⫾ 0.78
1.16 ⫾ 0.07 5.28 ⫾ 0.76 a 0.81 ⫾ 0.09 2.12 ⫾ 0.73 b
6.8 ⫾ 0.8 1281 ⫾ 301 a 7.7 ⫾ 0.4 85 ⫾ 47 b
Note. Values are means ⫾ SE for rats 12 h after treatment with saline (control) or Cd (3 mg/kg, sc). a Significantly different from age-matched controls, P ⬍ 0.05. b Significantly different from young-adults rats with the same treatment, P ⬍ 0.05.
72
YAMANO, DECICCO, AND RIKANS
TABLE 2 Effect of Gadolinium Chloride and Cyclosporin A on Cd-Induced Elevations of Hepatic IL-1b and CINC Treatment
Hepatic IL-1 (ng/g)
Hepatic CINC (ng/g)
Control Cd Gadolinium Gadolinium ⫹ Cd Cyclosporin A Cyclosporin A ⫹ Cd
2.48 ⫾ 0.24 3.59 ⫾ 0.26 a 2.27 ⫾ 0.13 2.53 ⫾ 0.16 b 2.63 ⫾ 0.16 2.68 ⫾ 0.17 b
1.59 ⫾ 0.22 4.74 ⫾ 0.70 a 1.75 ⫾ 0.11 2.41 ⫾ 0.21 a,b 1.34 ⫾ 0.42 3.25 ⫾ 0.43 a
Note. Values are means ⫾ SE for 5 to 6 rats 12 h after treatment. a Significantly different from rats receiving no Cd, P ⬍ 0.05. b Significantly different from treatment with Cd only, P ⬍ 0.05.
FIG. 4. Effect of cyclosporin A and gadolinium chloride pretreatment on Cd-induced hepatotoxicity. Rats were treated with saline (Cont) or CdCl 2 (Cd; 3 mg/kg sc) with and without pretreated with cyclosporin A (CsA; 5 mg/kg im) or gadolinium chloride (Gd; 10 m/kg iv). The rats were killed after 24 h, and serum enzyme activities were determined spectrophotometrically. Values are means ⫾ SE for five to six rats. **Significantly different, p ⬍ 0.01. NS, not significant.
events involving several types of liver cells and a large number of inflammatory and toxic mediators (Kayama et al., 1995; Sauer et al., 1997; Dong et al., 1998). The major participants in this process are Kupffer cells, neutrophils, and monocytes. As the cells of the immune system are affected in many ways by the process of aging (reviewed by Miller, 1996; Horan and Ashcroft, 1997), it seemed reasonable to propose that the influence of aging on the acute hepatotoxicity of Cd involves an alteration in the inflammatory process. The measurement of carbon clearance has been used to evaluate the phagocytic activity of Kupffer cells in vivo (Halpern et al., 1953; Szabo, 1983; Sauer et al., 1997) and in isolated perfused liver (Tapia et al., 1998). The procedure measures the removal of infused particles of colloidal carbon from the bloodstream by the reticuloendothelial system (RES). The Kupffer cells of the liver represent a major part of the RES and are primarily responsible for the clearance of harmful
substances from the circulation. In rats injected with 160 mg/kg of carbon (the dose employed in this study), about 90% of the carbon was found in hepatic Kupffer cells (Halpern et al., 1953). In the present study carbon clearance was used to evaluate Kupffer cell activation by Cd. The removal of carbon particles from the blood appeared to occur in two phases (fast and slow) and Cd profoundly affected the initial, fast phase of clearance. Pretreatment with gadolinium, an agent that inactivates macrophages, prevented the escalation by Cd of the initial phase of carbon clearance, confirming that the increase in carbon removal was due to the stimulation of Kupffer cell phagocytosis. The fact that the slow phase of carbon clearance was not stimulated by Cd suggests that a specific pathway of clearance was altered. Although a previous study had demonstrated an increase in carbon clearance 24 h after treatment
FIG. 5. Correlation between hepatic CINC levels 12 h after Cd administration and Cd-induced hepatotoxicity. Rats were treated with saline (Cont), Cd, gadolinium chloride (Gd), gadolinium chloride followed by Cd (Gd ⫹ Cd), cyclosporin A (CsA) and cyclosporin A followed by Cd (CsA ⫹ Cd). The rats were killed after 12 h, and hepatic CINC concentrations were determined by ELISA. In a parallel experiment, rats received the same treatments but were killed after 24 h, and SDH activities were determined spectrophotometrically. Serum SDH activity is plotted against hepatic CINC concentration for each treatment group (n ⫽ 5 to 6).
Cd HEPATOTOXICITY IN OLD AGE: ROLE OF IMMUNE RESPONSE
with Cd (Sauer et al., 1997), our results showed that Cd stimulated carbon clearance as early as 1.5 h after treatment. This activation of Kupffer cells at an early time point is consistent with a causal role for Kupffer cell activation in the development of inflammatory damage from Cd. The stimulation by Cd of the initial phase of carbon clearance was much greater in young-adult rats than in old rats, suggesting that the ability to activate hepatic Kupffer cells is markedly diminished in old age. These results for old rats are consistent with previous findings demonstrating an age-related decrement in the phagocytic activity of endotoxin-stimulated Kupffer cells (Durham et al., 1990). In addition, the results support other findings indicating that macrophage activation is impaired in old rats (Kauffman, 1986; Bradley et al., 1989; Davila et al., 1990; Brouwer et al., 1995). Inflammatory cells participate in the pathogenesis of tissue injury by releasing many products, some of which are mediators of the inflammatory response and some of which are cytotoxic. Several proinflammatory cytokines are secreted, e.g., TNF-␣, IL-1␣, IL-1, IL-6, and interferon-␥; however, TNF-␣ and IL-1 are recognized as the critical early mediators of tissue injury. In the present study we found that acute Cd hepatotoxicity was associated with increased concentrations of hepatic IL-1, but not TNF-␣ or IL-6. This result for TNF-␣ contrasts with results reported by Kayama et al. (1995), who found that cultured liver slices for Cd-exposed rats secreted increased amounts of TNF-␣ and IL-1␣ (IL-1 was not measured), but not IL-6. The apparent discrepancy in results for TNF-␣ might be due to the difference in methods used for cytokine determination; they measured cytokines in culture medium using bioassay whereas we measured cytokines in liver fractions using ELISA. A role for IL-1 in the hepatotoxicity produced by Cd has not been demonstrated previously; however, IL-1 has been implicated as a mediator of lipopolysaccharide toxicity (Alexander et al., 1992; Boermeester et al., 1995), and increased quantities of mRNA transcripts for IL-1 have been detected in liver slices of mice injected with Cd (Kayama et al., 1995). In addition to proinflammatory cytokines like TNF-␣ and IL-1, other cytokines known as chemokines also participate in the inflammatory process. The major effect of chemokines is to stimulate the chemotaxis of leukocytes, and the CXC family of chemokines selectively induces the migration of neutrophils, as opposed to monocytes (reviewed by Thompson et al., 1996). CXC chemokines are produced by several cell types of the liver in response to lipopolysaccharide, TNF-␣, or IL-1, although IL-1 appears to be the primary stimulus (Mawet et al., 1996; Anthony et al., 1998). The major CXC chemokine in rats is cytokine-induced neutrophil chemoattractant, or CINC. It was first identified in the medium of IL-1-stimulated rat glomerular epithelial cells and found to be functionally analogous to human IL-8 (Watanabe et al., 1989). As neutrophil infiltration into the site of injury is a prominent feature of Cd hepatotoxicity, our finding that CINC concentrations increased
73
dramatically in serum and liver of rats treated with Cd is not surprising. The Cd-induced elevation of hepatic CINC was preceded by an increase in IL-1, which is consistent with IL-1 acting as a stimulus for CINC production. Our results showing that Cd increased hepatic CINC concentrations in vivo extend previous findings in which Cd treatment in vitro stimulated the production of CINC by isolated rat hepatocytes (Dong et al., 1998). In addition, our results suggest that the severity of Cd-induced liver damage correlates directly with the amount of CINC present in the liver. Although the relative contributions of various cell types in the liver to the increased CINC concentrations observed after Cd treatment are not clear from our study, the fact that CINC levels were markedly reduced by gadolinium suggests that Kupffer cells play a major role in the production of CINC in vivo. The attenuation of Cd-induced liver injury in senescent rats compared with young-adult rats was associated with a substantial reduction in the production of hepatic CINC in the old rats. In addition, the rise in hepatic IL-1 produced by Cd was statistically significant in young-adult rats but not old ones; however, the apparent lack of a significant IL-1 response might have been due to variability in the data for old rats rather than an effect of aging on Cd-induced IL-1 production. Thus, another explanation for the results may be that the ability to produce CINC in response to IL-1 decreases as a function of age, as suggested by the studies of Anthony et al. (1998). It also is possible that CINC production in livers of Cd-treated rats is determined predominantly by factors other than IL-1 (e.g., by direct stimulation from Cd). The fact that the severity of liver damage produced by Cd in different age and treatment groups correlated with hepatic levels of CINC but not with those of IL-1 is consistent with either one of these possibilities. Additional support for the involvement of IL-1 and CINC in the inflammatory liver damage produced by Cd was provided by experiments utilizing inhibitors of the immune response. Cd-induced hepatotoxicity was markedly diminished by pretreatment with gadolinium chloride or cyclosporin A, and the reduction in severity of liver damage was associated with decreased levels of both IL-1 and CINC. These results are consistent with a role for both cytokines in the inflammatory damage produced by Cd. Gadolinium destroys Kupffer cells and prevents the release of inflammatory mediators; its prevention of Cd-induced acute hepatotoxicity has been demonstrated previously (Sauer et al., 1997). Cyclosporin A acts by inhibiting the activation of nuclear factor-B (Liu, 1993), a crucial regulator of the expression of several inflammatory cytokines including TNF-␣, IL-1, and CINC. A recent study showed that cyclosporin A diminished CINC production by Kupffer cells and attenuated neutrophil accumulation and ischemia/reperfusion damage in rat liver (Matsuda et al., 1998). The current findings are consistent with these results, but they do not exclude the possibility that other inflammatory media-
74
YAMANO, DECICCO, AND RIKANS
tors, in addition to IL-1 and CINC, contribute to Cd-induced liver damage. In summary, the results from this study demonstrated that concentrations of IL-1 and CINC are elevated in livers of rats administered a hepatotoxic dose of Cd. Pretreatment with gadolinium chloride or cyclosporin A decreased the Cd-induced increases in these cytokines and abrogated the hepatotoxicity. These results suggest that IL-1 and CINC are important mediators of the inflammatory liver damage produced by Cd. In addition, we found that Cd-mediated stimulation of Kupffer cell activity and elevations of hepatic CINC were markedly diminished in old rats compared with young-adult rats, suggesting that the attenuation of Cd-induced liver injury that occurs in old age is caused by an impairment in Kupffer cell activation, leading to lower production of CINC and less inflammatory liver injury. The mechanisms by which Kupffer cells and other cells are stimulated by Cd to initiate the inflammatory response are poorly understood. Furthermore, the inflammatory process involves several cascading events that are modulated by paracrine and autocrine pathways, and a small change at any step might produce a large difference in the final outcome. Further study is needed to fully elucidate the mechanism of the age effect on the inflammatory liver injury produced by Cd.
rat—Methods and standards. In The Rat in Laboratory Investigation (E. J. Farris and J. Q., Griffith, Jr., Eds.), pp. 406 – 420. Hafner Publishing, New York. Davila, D. R., Edwards, III, C. K., Arkins, S., Simon, J., and Kelley, K. W. (1990). Interferon-␥-induced priming for secretion of superoxide anion and tumor necrosis factor-␣ declines in macrophages from aged rats. FASEB J. 4, 2906 –2911. Dong, W., Simeonova, P. P., Gallucci, R., Matheson, J., and Flood, L., Wang, S., Hubbs, A., and Luster, M. I. (1998). Toxic metals stimulate inflammatory cytokines in hepatocytes through oxidative stress mechanisms. Toxicol. Appl. Pharmacol. 151, 359 –366. Durham, S. K., Brouwer, A., Barelds, R. J., Horan, M. A., and Knook, D. L. (1990). Comparative endotoxin-induced hepatic injury in young and aged rats. J. Pathol. 162, 341–349. Halpern, B. N., Benacerraf, B., and Biozzi, G. (1953). Quantitative study of the granulopectic activity of the reticulo-endothelial system. Br. J. Exp. Pathol. 34, 426 – 440. Hoffmann, E. O., Cook, J. A., Di Luzio, N. R., and Coover, J. A. (1975). The effects of acute cadmium administration in the liver and kidney of the rat: Light and electron microscopic studies. Lab. Invest. 32, 655– 664. Horan, M. A., and Ashcroft, G. S. (1997). Ageing, defense mechanisms and the immune system. Age Ageing 26-S4, 15–19. Jaeschke, H., and Smith, C. W. (1997). Mechanism of neutrophil-induced parenchymal cell injury. J. Leukocyte Biol. 61, 647– 653. Kauffman, C. A. (1986). Endogenous pyrogen/interleukin-1 production in aged rats. J. Gerontol. 21, 75–78.
ACKNOWLEDGMENTS
Kayama, F., Yoshida, T., Elwell, M. R., and Luster, M. I. (1995). Role of tumor necrosis factor-␣ in cadmium-induced hepatotoxicity. Toxicol. Appl. Pharmacol. 131, 224 –234.
This work was supported in part by a grant from the Presbyterian Health Foundation and by the Osaka City Institute of Public Health and Environmental Sciences.
Koizumi, T., Yokota, T., Shirakura, H., Tatsumoto, H., and Suzuki, K. T. (1994). Potential mechanism of cadmium-induced cytotoxicity in rat hepatocytes: Inhibitory action of cadmium on mitochondrial respiratory activity. Toxicology 92, 115–125.
REFERENCES Alexander, H. R., Doherty, G. M., Venzon, D. J., Merino, M. J., Fraker, D. L., and Norton, J. A. (1992). Recombinant interleukin-1 receptor antagonist (IL-1ra): Effective therapy against gram-negative sepsis in rats. Surgery 112, 188 –194. Anthony, D., Dempster, R., Fearn, S., Clements, J., Wells, G., Perry, V. H., and Walker, K. (1998). CXC chemokines generate age-related increases in neutrophil-mediated brain inflammation and blood– brain barrier breakdown. Curr. Biol. 8, 923–926. Badger, D. A., Kuester, R. K., Sauer, J. M., and Sipes, I. G. (1997). Gadolinium chloride reduces cytochrome P450: Relevance to chemical-induced hepatotoxicity. Toxicology 121, 143–153. Baggiolini, M., Dewald, B., and Moser, B. (1994). Interleukin-8 and related chemotactic cytokine-CXC and CC chemokines. Adv. Immunol. 55, 97–179. Boermeester, M. A., Straatsburg, I. H., Houdijk, A. P. J., Meyer, C., Frederiks, W. M., Wesdorp, R. I. C., van Noorden, C. J. F., and van Leeuwen, P. A. M. (1995). Endotoxin and interleukin-1 related hepatic inflammatory response promotes liver failure after partial hepatectomy. Hepatology 22, 1499 –1506. Bradley, S. F., Vibhagool, A., Kunkel, S. L., and Kauffman, C. A. (1989). Monokine secretion in aging and protein malnutrition. J. Leukocyte Biol. 45, 510 –514. Brouwer, A., Parker, S. G., Hendriks, H. F., Gibbons, L., and Horan, M. A. (1995). Production of eicosanoids and cytokines by Kupffer cells from young and old rats stimulated by endotoxin. Clin. Sci. 88, 211–217. Creskoff, A. J., Fitz-Hugh, T., Jr., and Farris, E. J. (1963). Hematology of the
Liu, J. (1993). FK-506 and cyclosporin, molecular probes for studying intracellular signal transduction. Immunol. Today 15, 274 –281. Matsuda, T., Yamaguchi, Y., Matsumura, F., Akizuki, E., Okabe, K., Liang, J., Ohshiro, H., Ichiguchi, O., Yamada, S., Mori, K., and Ogawa, M. (1998). Immunosuppressants decrease neutrophil chemoattractant and attenuate ischemia/reperfusion injury of the liver in rats. J. Trauma 44, 475– 484. Mawet, E., Shiratori, Y., Hikiba, Y., Takada, H., Yoshida, H., Okano, K., Komatsu, Y., Matsumura, M., Niwa, Y., and Omata, M. (1996). Cytokineinduced neutrophil chemoattractant release from hepatocytes is modulated by Kupffer cells. Hepatology 23, 353–358. Miller, R. A. (1996). The aging immune system: Primer and prospectus. Science 273, 70 –74. Mu¨ller, L. (1986). Consequences of cadmium toxicity in rat hepatocytes: Mitochondrial dysfunction and lipid peroxidation. Toxicology 40, 285–295. Ohkubo, K., Masumoto, T., Horiike, N., and Onji, M. (1998). Induction of CINC (interleukin-8) production in rat liver by non-parenchymal cells. J. Gastroenterol. Hepatol. 13, 696 –702. Rikans, L. E., Lopez, T. R., and Hornbrook, K. R. (1996). Age and gender differences in hepatic ascorbic acid concentrations and NADPH-dehydroascorbic acid reductase activity. Mech. Ageing Dev. 91, 165–169. Sauer, J. M., Waalkes, M. P., Hooser, S. B., Kuester, R. K., McQueen, C. A., and Sipes, I. G. (1997). Suppression of Kupffer cell function prevents cadmium induced hepatocellular necrosis in the Sprague–Dawley rat. Toxicology 121, 155–164. Stowe, H. D., Wilson, M., and Goyer, R. A. (1972). Clinical and morphological effects of oral cadmium toxicity in rabbits. Arch. Pathol. 94, 389 – 405. Strubelt, O., Kremer, J., Tilse, A., Keogh, J., Pentz, R., and Younes, M. (1996).
Cd HEPATOTOXICITY IN OLD AGE: ROLE OF IMMUNE RESPONSE Comparative studies on the toxicity of mercury, cadmium, and copper toward the isolated perfused rat liver. J. Toxicol. Environ. Health 47, 267–283. Szabo, S. (1983). Somatostatin stimulates clearance and hepatic uptake of colloidal carbon in the rat. Life Sci. 33, 1975–1980. Tapia, G., Troncoso, P., Galleano, M., Fernandez, V., Puntarulo, S., and Viedela, L. A. (1998). Time course study of the influence of acute iron overload on Kupffer cell functioning and hepatoxicity assessed in the isolated perfused rat liver. Hepatology 27, 1311–1316.
75
Thompson, K., Wright, S., and Sheron, N. (1996). Chemotactic and suppressor cytokine networks. Hepato-Gastroenterology 43, 15–31. Watanabe, K., Kinoshita, S., and Nakagawa, H. (1989). Purification and characterization of cytokine-induced neutrophil chemoattractant produced by epithelioid cell line of normal rat kidney (NRK-52E cell). Biochem. Biophys. Res. Commun. 161, 1093–1099. Yamano, T., Shimizu, M., and Noda, T. (1998). Age-related change in cadmium-induced hepatotoxicity in Wistar rats: Role of Kupffer cells and neutrophils. Toxicol. Appl. Pharmacol. 151, 9 –15.
Attenuation of Cadmium-Induced Liver Injury in Senescent Male Fischer 344 Rats: Role of Kupffer Cells and Inflammatory Cytokines Tetsuo Yamano,* Lacinda A. DeCicco,† and Lora E. Rikans† *Osaka City Institute of Public Health and Environmental Sciences, Osaka, Japan; and †Department of Pharmaceutical Sciences, University of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma 26901 Received June 28, 1999; accepted October 21, 1999
information relevant to metal toxicity as well as to the development of liver damage from other toxicants. Thus, Cd has become an important experimental hepatotoxicant. In a previous study we found that senescent rats are much more resistant to Cd-induced acute hepatotoxicity than are their younger counterparts and that age-associated changes in hepatic metallothionein and glutathione do not explain this phenomenon. These results are consistent with previous findings in which age-associated differences in these protective factors did not account for increased liver damage from Cd in 6-month-old rats compared with 1- or 2-month-old rats (Yamano et al., 1998). Instead, the results suggested that the age-dependent shift in sensitivity was due to changes in Kupffer cell and/or neutrophil activity. Thus, the current study was undertaken to examine the role of inflammatory processes in the postmaturational effects of aging on Cd hepatotoxicity. The involvement of Kupffer cells in Cd-mediated liver damage is suggested by the results of several investigations. Kupffer cell activation, demonstrated as increased cytoplasmic vacuolization or increased carbon clearance, has been observed after Cd administration to rats (Hoffman et al., 1975; Sauer et al., 1997). In addition, the administration of gadolinium chloride, an inhibitor of Kupffer cell function, alleviates acute Cd hepatotoxicity in vivo without substantially affecting hepatic Cd accumulation or hepatic metallothionein and glutathione contents (Sauer et al., 1997; Yamano et al., 1998). Inhibition by gadolinium was not observed in isolated hepatocytes, indicating that the prevention of Cd hepatotoxicity in vivo depended on the participation of nonparenchymal cells, i.e., Kupffer cells (Badger et al., 1997). Although the mechanisms by which Kupffer cells contribute to Cd-induced hepatotoxicity are not entirely clear, it is known that activated Kupffer cells release a number of inflammatory mediators (e.g., cytokines, chemokines, adhesion molecules) that initiate a cascade of cellular and humoral responses leading to inflammation and secondary damage to the liver. Infiltration of neutrophils at the site of injury is a common observation in Cd-induced hepatic necrosis (Stowe et al., 1972). The migration and activation of neutrophils are regulated by cytokines, chemokines, and adhesion molecules (Bag-
Attenuation of Cadmium-Induced Liver Injury in Senescent Male Fischer 344 Rats: Role of Kupffer Cells and Inflammatory Cytokines. Yamano, T., DeCicco, L. A., and Rikans, L. E. (2000). Toxicol. Appl. Pharmacol. 162, 68 –75. In the previous study we showed that senescent male Fischer 344 rats were resistant to Cd-induced hepatotoxicity compared with young-adult rats. In the present study we investigated the role of Kupffer cells and inflammatory cytokines in this effect of aging. The phagocytic activity of Kupffer cells, determined as the removal of carbon from blood, was stimulated by the administration of a hepatotoxic dose of Cd (3 mg/kg sc) in young-adult (5 months) rats but not in old (28 months) rats. Hepatic concentrations of interleukin (IL)-1 and cytokine-induced neutrophil chemoattractant (CINC), but not of tumor necrosis factor-␣ or IL-6, were elevated in young rats treated with Cd. In old rats, however, the increase in IL-1 produced by Cd was not statistically significant and the increase in CINC was much lower than in youngadult rats. Pretreatment with gadolinium chloride or cyclosporin A inhibited the elevations in hepatic cytokines and attenuated Cd-induced liver damage, assessed on the basis of serum alanine aminotransferase and sorbitol dehydrogenase activities. Cd-induced hepatotoxicity in the different treatment groups correlated well with hepatic levels of CINC (r ⴝ 0.98, p < 0.001) but not with those of IL-1. The results suggest that (1) Kupffer cell activation is essential for inflammatory liver damage from Cd, (2) IL-1 and CINC are important mediators of the inflammatory response induced by Cd, and (3) the attenuation of Cd-induced liver injury in senescent rats is caused by an impairment in Kupffer cell activation, leading to a lower production of CINC and less inflammatory liver injury. © 2000 Academic Press Key Words: Cadmium; cyclosporin A; gadolinium chloride; cytokine-induced neutrophil chemoattractant; interleukin IL-1; inflammatory liver injury; Kupffer cells.
Toxicity from chronic exposure to low doses of Cd is manifested mainly as chronic renal tubular disease and chronic pulmonary disease. On the other hand, acute exposure to Cd results primarily in liver damage, because the metal initially accumulates in the liver. Studies of Cd-induced hepatotoxicity in laboratory animals have contributed valuable mechanistic 0041-008X/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
68
Cd HEPATOTOXICITY IN OLD AGE: ROLE OF IMMUNE RESPONSE
giolini et al., 1994; Jaeschke and Smith, 1997; Ohkubo et al., 1998). Increased expression of several of these mediators, including tumor necrosis factor (TNF)-␣, interleukin (IL)-1␣, IL-1, IL-6, interferon-␥, intercellular adhesion molecule-1, macrophage inflammatory protein-2, and cytokine-induced neutrophil chemoattractant (CINC), has been demonstrated in rodent liver or hepatocytes exposed to Cd (Kayama et al., 1995; Dong et al., 1998). Moreover, pretreatment of mice with anti-TNF-␣ antibodies prevented Cd-induced secretion of acute phase proteins and focal inflammation in the liver (Kayama et al., 1995). These results indicate that inflammatory cytokines are expressed following Cd exposure and are responsible for the manifestations of inflammation observed in Cdinduced hepatotoxicity. In a previous study, we showed that the difference between young and mature rats in sensitivity to Cd-induced liver damage was decreased substantially by the inactivation of Kupffer cells or the depletion of neutrophils (Yamano et al., 1998). However, little is known about the effects of postmaturational aging on hepatotoxicant-induced inflammatory responses, such as Kupffer cell activation, cytokine production, or migration of neutrophils to the site of injury. In this study we compared rates of carbon clearance, a measure of Kupffer cell activity, and hepatic levels of cytokines in young-adult and old rats treated with Cd. METHODS Animals. Male Fischer 344 (F344) rats were obtained from the colony supported by the National Institute on Aging at Harlan Sprague Dawley Inc. (Indianapolis, IN). Young-adult rats were 5 months and old rats were 25 to 28 months of age. Initial experiments to determine the time course of cytokine elevations in liver and experiments with inhibitors were performed using younger rats (F344 males about 3 months old). The rats were acclimated to our facilities for 1 to 2 weeks before use, and they were fed and housed as described previously (Rikans et al., 1996). Treatments. Rats were administered saline or CdCl 2 (3 mg Cd/kg body wt) dissolved in saline as sc injections (2.5 ml/kg). Food was withdrawn after treatment. In some studies the rats were pretreated with either cyclosporin A (Sigma Chemical Co., St. Louis, MO) or gadolinium chloride. Cyclosporin A (5 mg/kg) was injected intramuscularly 24 and 3 h before treatment with Cd or saline (Matsuda et al., 1998), and gadolinium chloride (10 mg/kg) was injected iv 24 h before treatment (Yamano et al., 1998). The rats were killed under pentobarbital anesthesia (30 mg/kg ip) for collection of blood and liver samples. The livers were frozen immediately in liquid nitrogen and kept at ⫺70°C for subsequent analysis. Colloidal carbon clearance. The clearance of colloidal carbon from the blood was determined as described by Szabo (1983) and Sauer et al. (1997) with slight modification. The carbon solution was prepared as follows: Commercial ink (Pelikan drawing ink A) was dialyzed 48 h against distilled water through a semipermeable membrane (12,000 MW cut-off). The dialysate was centrifuged at 5000 rpm and the carbon content in the supernatant was determined by weight. The supernatant was diluted with distilled water and NaCl was added to make final concentrations equal to 50 mg/ml of carbon and 0.85% NaCl. Rats were administered CdCl 2 or saline with or without gadolinium chloride pretreatment, as described above. Carbon solution (160 mg/kg) was injected into the tail vein under pentobarbital anesthesia (50 mg/kg ip) at 0.5, 1.5, 3, or 6 h after treatment with Cd or saline. Blood samples were
69
collected from the other side of the tail vein just before carbon injection and at 3, 6, 12, and 24 min after carbon injection. Each sample (44.7 l, collected in heparinized capillary tubes) was added to 2 ml of a 0.1% Na 2CO 3 solution and the absorbance of the hemolyzed blood was determined at 640 nm against a blank containing blood drawn prior to carbon injection. Carbon concentrations in blood were calculated from a standard curve of known carbon concentrations. Serum activities of hepatic enzymes. Serum alanine aminotransferase (ALT) and sorbitol dehydrogenase (SDH) activities were determined spectrophotometrically, using diagnostic kits from Sigma Chemical Co. Cytokine assays. Portions of liver were homogenized in nine parts of phosphate-buffered saline (PBS), pH 7.4, containing 2 mM of phenylmethylsulfonyl fluoride and 1 g/ml each of antipain, leupeptin, and pepstain A (all from Sigma Chemical Co.). The homogenates were centrifuged for 15 min at 15,000g at 8°C, and the supernatants were removed and stored at ⫺70°C until the assays were performed. Cytokine concentrations were determined using enzyme-linked immunosorbent assay (ELISA) procedures. TNF-␣ and IL-6 were measured using Cytoscreen ELISA kits from Biosource International (Camarillo, CA), and IL-1 was determined by a rat ELISA kit from Endogen (Woburn, MA). CINC concentrations also were quantified by ELISA. Microtiter plates were coated overnight at 4°C with polyclonal rabbit anti-rat CINC antibody (Pepro Tech Inc., Rocky Hill, NJ), using 100 l/well of a 1000 ng/ml solution in carbonate buffer, pH 9.6. After washing three times with PBS containing 0.05% Tween-20, blocking solution (PBS containing 1% BSA, 300 l/well) was added. The plates were sealed, incubated 1 h at 37°C, and washed. Recombinant rat CINC (Pepro Tech Inc.) or samples to be analyzed, diluted appropriately with PBS containing 0.1% BSA and 0.05% Tween-20, were added (100 l/well). The plates were sealed and incubated for 1 h at room temperature. After washing, biotinylated polyclonal goat anti-rat CINC antibody (R & D systems, Minneapolis, MN) in PBS containing 0.1% BSA and 0.05% Tween-20 was added (100 ng/ml, 100 l/well), and the plates were sealed and incubated for 1 h at room temperature. Following another washing, horseradish peroxidase-conjugated streptavidin (Pierce, Rockford, IL) was added (100 ng/ml, 100 l/well), and the plates were sealed and incubated for 20 min at room temperature. The plates were washed and incubated for 15 to 30 min in the dark with 100 l/well of o-phenylenediamine dihydrochloride buffer solution for ELISA (Sigma Chemical Co.). The enzymatic reaction was stopped by the addition of 50 l of 4 N HCl and the plates were read at 490 nm. The concentrations of CINC in liver fractions were calculated from a standard curve. The assay was linear for CINC concentrations between 7.8 and 500 pg/ml (0.4 to 25 ng/g liver), and the recovery of added recombinant rat CINC from liver fractions ranged from 89 to 112%. Statistics. Results are expressed as means ⫾ SE. Data were analyzed by Student’s t test, the Mann–Whitney rank sum test, one-way analysis of variance followed by Tukey’s test, or the Kruskal–Wallis test as appropriate.
RESULTS
The influence of Cd on the phagocytic activity of Kupffer cells in vivo was assessed by determining the removal of colloidal carbon from the blood. Carbon clearance in Cdtreated vs. control rats was significantly increased 1.5 h after Cd administration, and the clearance became progressively greater after 3 and 6 h. Because Cd-induced liver damage, indicated by elevated serum SDH levels, was evident by 6 h, all subsequent measurements of carbon clearance were determined 3 h after Cd administration. Figure 1 shows carbon concentrations in blood at 3, 6, 12, and 24 min after carbon injection. The value shown at 0 min is a theoretical initial concentration calculated by dividing the dose of carbon administered (160 mg/kg body wt) by the blood volume (67 ml/kg
70
YAMANO, DECICCO, AND RIKANS
FIG. 1. Effects of Cd and gadolinium chloride on carbon clearance. Young male F344 rats were treated with saline (Cont) or CdCl 2 (Cd; 3 mg/kg sc) 3 h prior to colloidal carbon injection (160 mg/kg iv), and blood carbon concentrations were determined at 3, 6, 12, and 24 min. Some rats were pretreated with gadolinium chloride (Gd; 10 mg/kg iv) 24 h prior to treatment with saline or Cd. Dotted lines denote estimated initial rates of clearance using a calculated value (2.4 mg/ml) for the initial carbon concentration. Values are means ⫾ SE for four to six rats. There are statistically significant differences in 3-min carbon concentrations between Cont vs. Cd ( p ⬍ 0.01), Cont vs Gd ( p ⬍ 0.05), and Cd vs Gd ⫹ Cd ( p ⬍ 0.001).
body wt) (Creskoff et al., 1949). There appeared to be two phases of carbon removal, an initial phase with a fast rate and a second phase with a slower rate. Initial clearance rates could not be calculated accurately because blood samples taken earlier than 3 min yielded erratic values for carbon concentration; however, Cd treatment caused a substantial decrease in 3-min carbon concentrations (0.84 ⫾ 0.09 vs. 1.42 ⫾ 0.05 in Cdtreated rats vs. saline-treated control rats, p ⬍ 0.01), suggesting that Cd treatment increased the initial rate of clearance. In contrast, pretreatment with gadolinium chloride increased the concentration of carbon remaining at 3 min (1.93 ⫾ 0.16 vs. 1.42 ⫾ 0.05 mg/ml in rats gadolinium-pretreated rats vs. control rats, p ⬍ 0.05, and 1.69 ⫾ 0.10 vs 0.84 ⫾ 0.09 mg/ml in rats receiving gadolinium ⫹ Cd vs. rats receiving Cd only, p ⬍ 0.001), suggesting a decrease in the initial clearance. The slow phase of carbon removal also was affected by Cd administration, but to a lesser extent. The effect of animal age on Kupffer cell activation was determined by comparing the effect of Cd on carbon removal in young-adult and old rats (5 and 28 months of age, respectively). The dose of Cd (3.0 mg/kg) used in these experiments was the same as the one used in the study demonstrating a marked attenuation of Cd-induced liver damage in old rats compared with young-adult rats. The amount of carbon remaining in the blood 3 min after carbon injection was similar in young-adult rats and old rats (1.89 ⫾ 0.15 and 1.67 ⫾ 0.17 mg/ml), suggesting that the initial phase of carbon clearance in control rats was unaffected by aging (Fig. 2). In contrast, the 3-min carbon levels were much lower in young-adult rats treated with Cd than in old rats treated with Cd (0.62 ⫾ 0.04 vs 1.56 ⫾ 0.14 mg/ml, p ⬍ 0.05), indicating that the stimula-
tion by Cd of carbon removal was more pronounced in youngadult rats than in old rats. In fact, no effect of Cd on carbon clearance was observed in the old animals. Concentrations of TNF-␣, IL-1, IL-6, and CINC were determined in liver and serum of young rats at various times after treatment with Cd or saline. Cd administration caused significant increases in hepatic levels of IL-1 and CINC (Fig. 3). A twofold elevation in IL-1 was apparent 6 h after Cd treatment. IL-1 was still elevated at the 12-h time point but returned to normal by 24 h. The rise in hepatic CINC concentration followed the rise in IL-1, beginning 12 h after Cd treatment and continuing for at least 24 h. Hepatic CINC levels at 24 h were three times greater in Cd-treated rats than in control rats. Serum SDH and ALT activities are also shown in Fig. 3, so that time courses for the increases in these cytokine levels can be compared with the development of hepatotoxic-
FIG. 2. Effect of Cd on carbon clearance in young-adult (5 months) and old (28 months) male F344 rats. The rats were treated with saline (Cont) or CdCl 2 (Cd; 3 mg Cd/kg sc) 3 h prior to colloidal carbon injection (160 mg/kg iv), and blood carbon concentrations were determined at 3, 6, 12, and 24 min. Dotted lines denote estimated initial rates of clearance using a calculated value (2.4 mg/ml) for the initial carbon concentration. Values are means ⫾ SE for four to six rats. There are statistically significant differences in 3-min carbon concentrations between young-adult Cont vs young-adult Cd ( p ⬍ 0.05) and young-adult Cd vs old Cd ( p ⬍ 0.05).
71
Cd HEPATOTOXICITY IN OLD AGE: ROLE OF IMMUNE RESPONSE
FIG. 3. Cd-mediated increases in hepatic IL-1 and CINC concentrations and in serum ALT and SDH activities. Young male F344 rats were treated with saline (Cont) or CdCl 2 (Cd; 3 mg/kg sc). IL-1 and CINC concentrations were measured by ELISA procedures, and serum activities of liver enzymes were measured spectrophotometrically. Values are means ⫾ SE for five to seven rats. *Effect of Cd treatment is significant, p ⬍ 0.05. **Effect of Cd treatment is significant, p ⬍ 0.01.
ity. Cd treatment did not affect hepatic concentrations of TNF-␣ or IL-6 (data not shown). The levels of TNF-␣, IL-1, and IL-6 in serum of control and Cd-treated rats were close to the limits of detection an did not correlate, either positively or negatively, with the levels in liver. In contrast, CINC levels in serum were easily detectable, and increases produced by Cd treatment paralleled those seen in liver (data not shown). Since hepatic concentrations of both IL-1 and CINC were elevated 12 h after Cd treatment, the effects of aging on Cd-mediated increases in the levels of these cytokines were determined at the 12-h time point. As expected, Il-1 levels were significantly increased in young-adult rats as a consequence of Cd treatment; however, the effect of Cd on IL-1 was not significant in old rats (Table 1). Likewise, hepatic CINC concentrations were 4.5-fold greater in Cd-treated young-adult rats than in control rats of the same age, whereas CINC levels in old rats were not increased significantly by Cd treatment. Serum SDH activities at 12 h were higher in youngadult rats than in old rats, confirming our earlier results that Cd-induced hepatotoxicity is significantly diminished in old age.
In order to explore the relationships between Cd-induced hepatotoxicity and Cd-mediated increases in hepatic IL-1 and CINC levels, rats were treated with gadolinium chloride, a Kupffer cell inhibitor, or cyclosporin A, an immunosuppressant, prior to Cd administration. Pretreatment with the inhibitors almost completely abrogated the hepatotoxicity, as shown by marked suppression of serum ALT and SDH activities 24 h after Cd treatment (Fig. 4). In addition, the inhibitors diminished the Cd-mediated elevations in hepatic IL-1 and CINC concentrations. The 12-h values for IL-1 and CINC are shown in Table 2. Gadolinium chloride completely suppressed the increase in IL-1 and partially suppressed the increase in CINC. Although cyclosporin A also suppressed IL-1 completely, its effect on CINC was not statistically significant at 12 h. The depression by inhibitors of hepatic CINC levels in Cd-treated rats was more pronounced at 24 h, and both inhibitors produced significant decreases in CINC at this time point (not shown). The 12-h concentrations of hepatic CINC in the various groups of rats correlated well with Cd-induced hepatotoxicity, measured as the leakage of SDH into the bloodstream (Fig. 5); however, correlations of IL-1 with hepatotoxicity were not significant. The severity of liver damage produced by Cd in young (3 months), young-adult (5 months), and old (25 to 28 months) rats also correlated well with hepatic CINC but not IL-1 levels (not shown). DISCUSSION
The mechanisms of acute Cd-mediated hepatotoxicity involves two pathways, one for the initial injury produced by direct effects of Cd and the other for the subsequent injury produced by the inflammatory response. Primary injury probably results from the binding of Cd to sulfhydryl groups on critical molecules in mitochondria, causing oxidative stress and the mitochondrial membrane permeability transition (Mu¨ller, 1986; Koizumi et al., 1994; Strubelt et al., 1996). Secondary injury from acute Cd exposure is thought to occur from the activation of Kupffer cells and a complex series of interactive TABLE 1 Effect of Age on Cd-Induced Elevations of Hepatic IL-1, Hepatic CINC, and Serum SDH Age (months)
Treatment
Hepatic IL-1 (ng/g)
Hepatic CINC (ng/g)
Serum SDH (IU/L)
5 5 25 25
Control Cd Conrol Cd
1.32 ⫾ 0.14 2.17 ⫾ 0.25 a 1.77 ⫾ 0.19 2.71 ⫾ 0.78
1.16 ⫾ 0.07 5.28 ⫾ 0.76 a 0.81 ⫾ 0.09 2.12 ⫾ 0.73 b
6.8 ⫾ 0.8 1281 ⫾ 301 a 7.7 ⫾ 0.4 85 ⫾ 47 b
Note. Values are means ⫾ SE for rats 12 h after treatment with saline (control) or Cd (3 mg/kg, sc). a Significantly different from age-matched controls, P ⬍ 0.05. b Significantly different from young-adults rats with the same treatment, P ⬍ 0.05.
72
YAMANO, DECICCO, AND RIKANS
TABLE 2 Effect of Gadolinium Chloride and Cyclosporin A on Cd-Induced Elevations of Hepatic IL-1b and CINC Treatment
Hepatic IL-1 (ng/g)
Hepatic CINC (ng/g)
Control Cd Gadolinium Gadolinium ⫹ Cd Cyclosporin A Cyclosporin A ⫹ Cd
2.48 ⫾ 0.24 3.59 ⫾ 0.26 a 2.27 ⫾ 0.13 2.53 ⫾ 0.16 b 2.63 ⫾ 0.16 2.68 ⫾ 0.17 b
1.59 ⫾ 0.22 4.74 ⫾ 0.70 a 1.75 ⫾ 0.11 2.41 ⫾ 0.21 a,b 1.34 ⫾ 0.42 3.25 ⫾ 0.43 a
Note. Values are means ⫾ SE for 5 to 6 rats 12 h after treatment. a Significantly different from rats receiving no Cd, P ⬍ 0.05. b Significantly different from treatment with Cd only, P ⬍ 0.05.
FIG. 4. Effect of cyclosporin A and gadolinium chloride pretreatment on Cd-induced hepatotoxicity. Rats were treated with saline (Cont) or CdCl 2 (Cd; 3 mg/kg sc) with and without pretreated with cyclosporin A (CsA; 5 mg/kg im) or gadolinium chloride (Gd; 10 m/kg iv). The rats were killed after 24 h, and serum enzyme activities were determined spectrophotometrically. Values are means ⫾ SE for five to six rats. **Significantly different, p ⬍ 0.01. NS, not significant.
events involving several types of liver cells and a large number of inflammatory and toxic mediators (Kayama et al., 1995; Sauer et al., 1997; Dong et al., 1998). The major participants in this process are Kupffer cells, neutrophils, and monocytes. As the cells of the immune system are affected in many ways by the process of aging (reviewed by Miller, 1996; Horan and Ashcroft, 1997), it seemed reasonable to propose that the influence of aging on the acute hepatotoxicity of Cd involves an alteration in the inflammatory process. The measurement of carbon clearance has been used to evaluate the phagocytic activity of Kupffer cells in vivo (Halpern et al., 1953; Szabo, 1983; Sauer et al., 1997) and in isolated perfused liver (Tapia et al., 1998). The procedure measures the removal of infused particles of colloidal carbon from the bloodstream by the reticuloendothelial system (RES). The Kupffer cells of the liver represent a major part of the RES and are primarily responsible for the clearance of harmful
substances from the circulation. In rats injected with 160 mg/kg of carbon (the dose employed in this study), about 90% of the carbon was found in hepatic Kupffer cells (Halpern et al., 1953). In the present study carbon clearance was used to evaluate Kupffer cell activation by Cd. The removal of carbon particles from the blood appeared to occur in two phases (fast and slow) and Cd profoundly affected the initial, fast phase of clearance. Pretreatment with gadolinium, an agent that inactivates macrophages, prevented the escalation by Cd of the initial phase of carbon clearance, confirming that the increase in carbon removal was due to the stimulation of Kupffer cell phagocytosis. The fact that the slow phase of carbon clearance was not stimulated by Cd suggests that a specific pathway of clearance was altered. Although a previous study had demonstrated an increase in carbon clearance 24 h after treatment
FIG. 5. Correlation between hepatic CINC levels 12 h after Cd administration and Cd-induced hepatotoxicity. Rats were treated with saline (Cont), Cd, gadolinium chloride (Gd), gadolinium chloride followed by Cd (Gd ⫹ Cd), cyclosporin A (CsA) and cyclosporin A followed by Cd (CsA ⫹ Cd). The rats were killed after 12 h, and hepatic CINC concentrations were determined by ELISA. In a parallel experiment, rats received the same treatments but were killed after 24 h, and SDH activities were determined spectrophotometrically. Serum SDH activity is plotted against hepatic CINC concentration for each treatment group (n ⫽ 5 to 6).
Cd HEPATOTOXICITY IN OLD AGE: ROLE OF IMMUNE RESPONSE
with Cd (Sauer et al., 1997), our results showed that Cd stimulated carbon clearance as early as 1.5 h after treatment. This activation of Kupffer cells at an early time point is consistent with a causal role for Kupffer cell activation in the development of inflammatory damage from Cd. The stimulation by Cd of the initial phase of carbon clearance was much greater in young-adult rats than in old rats, suggesting that the ability to activate hepatic Kupffer cells is markedly diminished in old age. These results for old rats are consistent with previous findings demonstrating an age-related decrement in the phagocytic activity of endotoxin-stimulated Kupffer cells (Durham et al., 1990). In addition, the results support other findings indicating that macrophage activation is impaired in old rats (Kauffman, 1986; Bradley et al., 1989; Davila et al., 1990; Brouwer et al., 1995). Inflammatory cells participate in the pathogenesis of tissue injury by releasing many products, some of which are mediators of the inflammatory response and some of which are cytotoxic. Several proinflammatory cytokines are secreted, e.g., TNF-␣, IL-1␣, IL-1, IL-6, and interferon-␥; however, TNF-␣ and IL-1 are recognized as the critical early mediators of tissue injury. In the present study we found that acute Cd hepatotoxicity was associated with increased concentrations of hepatic IL-1, but not TNF-␣ or IL-6. This result for TNF-␣ contrasts with results reported by Kayama et al. (1995), who found that cultured liver slices for Cd-exposed rats secreted increased amounts of TNF-␣ and IL-1␣ (IL-1 was not measured), but not IL-6. The apparent discrepancy in results for TNF-␣ might be due to the difference in methods used for cytokine determination; they measured cytokines in culture medium using bioassay whereas we measured cytokines in liver fractions using ELISA. A role for IL-1 in the hepatotoxicity produced by Cd has not been demonstrated previously; however, IL-1 has been implicated as a mediator of lipopolysaccharide toxicity (Alexander et al., 1992; Boermeester et al., 1995), and increased quantities of mRNA transcripts for IL-1 have been detected in liver slices of mice injected with Cd (Kayama et al., 1995). In addition to proinflammatory cytokines like TNF-␣ and IL-1, other cytokines known as chemokines also participate in the inflammatory process. The major effect of chemokines is to stimulate the chemotaxis of leukocytes, and the CXC family of chemokines selectively induces the migration of neutrophils, as opposed to monocytes (reviewed by Thompson et al., 1996). CXC chemokines are produced by several cell types of the liver in response to lipopolysaccharide, TNF-␣, or IL-1, although IL-1 appears to be the primary stimulus (Mawet et al., 1996; Anthony et al., 1998). The major CXC chemokine in rats is cytokine-induced neutrophil chemoattractant, or CINC. It was first identified in the medium of IL-1-stimulated rat glomerular epithelial cells and found to be functionally analogous to human IL-8 (Watanabe et al., 1989). As neutrophil infiltration into the site of injury is a prominent feature of Cd hepatotoxicity, our finding that CINC concentrations increased
73
dramatically in serum and liver of rats treated with Cd is not surprising. The Cd-induced elevation of hepatic CINC was preceded by an increase in IL-1, which is consistent with IL-1 acting as a stimulus for CINC production. Our results showing that Cd increased hepatic CINC concentrations in vivo extend previous findings in which Cd treatment in vitro stimulated the production of CINC by isolated rat hepatocytes (Dong et al., 1998). In addition, our results suggest that the severity of Cd-induced liver damage correlates directly with the amount of CINC present in the liver. Although the relative contributions of various cell types in the liver to the increased CINC concentrations observed after Cd treatment are not clear from our study, the fact that CINC levels were markedly reduced by gadolinium suggests that Kupffer cells play a major role in the production of CINC in vivo. The attenuation of Cd-induced liver injury in senescent rats compared with young-adult rats was associated with a substantial reduction in the production of hepatic CINC in the old rats. In addition, the rise in hepatic IL-1 produced by Cd was statistically significant in young-adult rats but not old ones; however, the apparent lack of a significant IL-1 response might have been due to variability in the data for old rats rather than an effect of aging on Cd-induced IL-1 production. Thus, another explanation for the results may be that the ability to produce CINC in response to IL-1 decreases as a function of age, as suggested by the studies of Anthony et al. (1998). It also is possible that CINC production in livers of Cd-treated rats is determined predominantly by factors other than IL-1 (e.g., by direct stimulation from Cd). The fact that the severity of liver damage produced by Cd in different age and treatment groups correlated with hepatic levels of CINC but not with those of IL-1 is consistent with either one of these possibilities. Additional support for the involvement of IL-1 and CINC in the inflammatory liver damage produced by Cd was provided by experiments utilizing inhibitors of the immune response. Cd-induced hepatotoxicity was markedly diminished by pretreatment with gadolinium chloride or cyclosporin A, and the reduction in severity of liver damage was associated with decreased levels of both IL-1 and CINC. These results are consistent with a role for both cytokines in the inflammatory damage produced by Cd. Gadolinium destroys Kupffer cells and prevents the release of inflammatory mediators; its prevention of Cd-induced acute hepatotoxicity has been demonstrated previously (Sauer et al., 1997). Cyclosporin A acts by inhibiting the activation of nuclear factor-B (Liu, 1993), a crucial regulator of the expression of several inflammatory cytokines including TNF-␣, IL-1, and CINC. A recent study showed that cyclosporin A diminished CINC production by Kupffer cells and attenuated neutrophil accumulation and ischemia/reperfusion damage in rat liver (Matsuda et al., 1998). The current findings are consistent with these results, but they do not exclude the possibility that other inflammatory media-
74
YAMANO, DECICCO, AND RIKANS
tors, in addition to IL-1 and CINC, contribute to Cd-induced liver damage. In summary, the results from this study demonstrated that concentrations of IL-1 and CINC are elevated in livers of rats administered a hepatotoxic dose of Cd. Pretreatment with gadolinium chloride or cyclosporin A decreased the Cd-induced increases in these cytokines and abrogated the hepatotoxicity. These results suggest that IL-1 and CINC are important mediators of the inflammatory liver damage produced by Cd. In addition, we found that Cd-mediated stimulation of Kupffer cell activity and elevations of hepatic CINC were markedly diminished in old rats compared with young-adult rats, suggesting that the attenuation of Cd-induced liver injury that occurs in old age is caused by an impairment in Kupffer cell activation, leading to lower production of CINC and less inflammatory liver injury. The mechanisms by which Kupffer cells and other cells are stimulated by Cd to initiate the inflammatory response are poorly understood. Furthermore, the inflammatory process involves several cascading events that are modulated by paracrine and autocrine pathways, and a small change at any step might produce a large difference in the final outcome. Further study is needed to fully elucidate the mechanism of the age effect on the inflammatory liver injury produced by Cd.
rat—Methods and standards. In The Rat in Laboratory Investigation (E. J. Farris and J. Q., Griffith, Jr., Eds.), pp. 406 – 420. Hafner Publishing, New York. Davila, D. R., Edwards, III, C. K., Arkins, S., Simon, J., and Kelley, K. W. (1990). Interferon-␥-induced priming for secretion of superoxide anion and tumor necrosis factor-␣ declines in macrophages from aged rats. FASEB J. 4, 2906 –2911. Dong, W., Simeonova, P. P., Gallucci, R., Matheson, J., and Flood, L., Wang, S., Hubbs, A., and Luster, M. I. (1998). Toxic metals stimulate inflammatory cytokines in hepatocytes through oxidative stress mechanisms. Toxicol. Appl. Pharmacol. 151, 359 –366. Durham, S. K., Brouwer, A., Barelds, R. J., Horan, M. A., and Knook, D. L. (1990). Comparative endotoxin-induced hepatic injury in young and aged rats. J. Pathol. 162, 341–349. Halpern, B. N., Benacerraf, B., and Biozzi, G. (1953). Quantitative study of the granulopectic activity of the reticulo-endothelial system. Br. J. Exp. Pathol. 34, 426 – 440. Hoffmann, E. O., Cook, J. A., Di Luzio, N. R., and Coover, J. A. (1975). The effects of acute cadmium administration in the liver and kidney of the rat: Light and electron microscopic studies. Lab. Invest. 32, 655– 664. Horan, M. A., and Ashcroft, G. S. (1997). Ageing, defense mechanisms and the immune system. Age Ageing 26-S4, 15–19. Jaeschke, H., and Smith, C. W. (1997). Mechanism of neutrophil-induced parenchymal cell injury. J. Leukocyte Biol. 61, 647– 653. Kauffman, C. A. (1986). Endogenous pyrogen/interleukin-1 production in aged rats. J. Gerontol. 21, 75–78.
ACKNOWLEDGMENTS
Kayama, F., Yoshida, T., Elwell, M. R., and Luster, M. I. (1995). Role of tumor necrosis factor-␣ in cadmium-induced hepatotoxicity. Toxicol. Appl. Pharmacol. 131, 224 –234.
This work was supported in part by a grant from the Presbyterian Health Foundation and by the Osaka City Institute of Public Health and Environmental Sciences.
Koizumi, T., Yokota, T., Shirakura, H., Tatsumoto, H., and Suzuki, K. T. (1994). Potential mechanism of cadmium-induced cytotoxicity in rat hepatocytes: Inhibitory action of cadmium on mitochondrial respiratory activity. Toxicology 92, 115–125.
REFERENCES Alexander, H. R., Doherty, G. M., Venzon, D. J., Merino, M. J., Fraker, D. L., and Norton, J. A. (1992). Recombinant interleukin-1 receptor antagonist (IL-1ra): Effective therapy against gram-negative sepsis in rats. Surgery 112, 188 –194. Anthony, D., Dempster, R., Fearn, S., Clements, J., Wells, G., Perry, V. H., and Walker, K. (1998). CXC chemokines generate age-related increases in neutrophil-mediated brain inflammation and blood– brain barrier breakdown. Curr. Biol. 8, 923–926. Badger, D. A., Kuester, R. K., Sauer, J. M., and Sipes, I. G. (1997). Gadolinium chloride reduces cytochrome P450: Relevance to chemical-induced hepatotoxicity. Toxicology 121, 143–153. Baggiolini, M., Dewald, B., and Moser, B. (1994). Interleukin-8 and related chemotactic cytokine-CXC and CC chemokines. Adv. Immunol. 55, 97–179. Boermeester, M. A., Straatsburg, I. H., Houdijk, A. P. J., Meyer, C., Frederiks, W. M., Wesdorp, R. I. C., van Noorden, C. J. F., and van Leeuwen, P. A. M. (1995). Endotoxin and interleukin-1 related hepatic inflammatory response promotes liver failure after partial hepatectomy. Hepatology 22, 1499 –1506. Bradley, S. F., Vibhagool, A., Kunkel, S. L., and Kauffman, C. A. (1989). Monokine secretion in aging and protein malnutrition. J. Leukocyte Biol. 45, 510 –514. Brouwer, A., Parker, S. G., Hendriks, H. F., Gibbons, L., and Horan, M. A. (1995). Production of eicosanoids and cytokines by Kupffer cells from young and old rats stimulated by endotoxin. Clin. Sci. 88, 211–217. Creskoff, A. J., Fitz-Hugh, T., Jr., and Farris, E. J. (1963). Hematology of the
Liu, J. (1993). FK-506 and cyclosporin, molecular probes for studying intracellular signal transduction. Immunol. Today 15, 274 –281. Matsuda, T., Yamaguchi, Y., Matsumura, F., Akizuki, E., Okabe, K., Liang, J., Ohshiro, H., Ichiguchi, O., Yamada, S., Mori, K., and Ogawa, M. (1998). Immunosuppressants decrease neutrophil chemoattractant and attenuate ischemia/reperfusion injury of the liver in rats. J. Trauma 44, 475– 484. Mawet, E., Shiratori, Y., Hikiba, Y., Takada, H., Yoshida, H., Okano, K., Komatsu, Y., Matsumura, M., Niwa, Y., and Omata, M. (1996). Cytokineinduced neutrophil chemoattractant release from hepatocytes is modulated by Kupffer cells. Hepatology 23, 353–358. Miller, R. A. (1996). The aging immune system: Primer and prospectus. Science 273, 70 –74. Mu¨ller, L. (1986). Consequences of cadmium toxicity in rat hepatocytes: Mitochondrial dysfunction and lipid peroxidation. Toxicology 40, 285–295. Ohkubo, K., Masumoto, T., Horiike, N., and Onji, M. (1998). Induction of CINC (interleukin-8) production in rat liver by non-parenchymal cells. J. Gastroenterol. Hepatol. 13, 696 –702. Rikans, L. E., Lopez, T. R., and Hornbrook, K. R. (1996). Age and gender differences in hepatic ascorbic acid concentrations and NADPH-dehydroascorbic acid reductase activity. Mech. Ageing Dev. 91, 165–169. Sauer, J. M., Waalkes, M. P., Hooser, S. B., Kuester, R. K., McQueen, C. A., and Sipes, I. G. (1997). Suppression of Kupffer cell function prevents cadmium induced hepatocellular necrosis in the Sprague–Dawley rat. Toxicology 121, 155–164. Stowe, H. D., Wilson, M., and Goyer, R. A. (1972). Clinical and morphological effects of oral cadmium toxicity in rabbits. Arch. Pathol. 94, 389 – 405. Strubelt, O., Kremer, J., Tilse, A., Keogh, J., Pentz, R., and Younes, M. (1996).
Cd HEPATOTOXICITY IN OLD AGE: ROLE OF IMMUNE RESPONSE Comparative studies on the toxicity of mercury, cadmium, and copper toward the isolated perfused rat liver. J. Toxicol. Environ. Health 47, 267–283. Szabo, S. (1983). Somatostatin stimulates clearance and hepatic uptake of colloidal carbon in the rat. Life Sci. 33, 1975–1980. Tapia, G., Troncoso, P., Galleano, M., Fernandez, V., Puntarulo, S., and Viedela, L. A. (1998). Time course study of the influence of acute iron overload on Kupffer cell functioning and hepatoxicity assessed in the isolated perfused rat liver. Hepatology 27, 1311–1316.
75
Thompson, K., Wright, S., and Sheron, N. (1996). Chemotactic and suppressor cytokine networks. Hepato-Gastroenterology 43, 15–31. Watanabe, K., Kinoshita, S., and Nakagawa, H. (1989). Purification and characterization of cytokine-induced neutrophil chemoattractant produced by epithelioid cell line of normal rat kidney (NRK-52E cell). Biochem. Biophys. Res. Commun. 161, 1093–1099. Yamano, T., Shimizu, M., and Noda, T. (1998). Age-related change in cadmium-induced hepatotoxicity in Wistar rats: Role of Kupffer cells and neutrophils. Toxicol. Appl. Pharmacol. 151, 9 –15.