- Email: [email protected]
Zinc-induced metallothionein and copper metabolism in intestinal mucosa, liver, and kidney of rats
Nutrition Research, Vol. 14, No. 6, pp. 897-908, 1994 Copyright 9 1994 Elsevie~ Science Lid Printed in the USA. All rights reserved 0271-5317/94 $6.00 + .00
Pergamon
0271-5317(94)E0023-W
Z I N C - I N D U C E D M E T A L L O T H I O N E I N AND C O P P E R M E T A B O L I S M IN INTESTINAL MUCOSA, LIVER, AND KIDNEY O F RATS 1 Philip G. Reeves 2, Kerry L. Rossow 3, and Dennis J. Bobilya 4
United States Department of Agriculture Agricultural Research Service Grand Forks Human Nutrition Research Center Grand Forks, ND 58202-9034 ABSTRACT
Large doses of parenteral zinc (Zn) and/or the feeding of high Zn diets to animals or humans for long periods affects copper (Cu) metabolism. Previous work suggests that Zninduced metallothionein (Mr) in intestinal epithelial ceils binds Cu and inhibits its absorption. This study was designed to determine the effects of treating rats with high dietary or high parenteral Zn on Cu metabolism and its relationship to MT in the intestinal epithelium, liver and kidney. Six-week-old male rats were fed for one week a control diet containing 42 nag Zn and 6 mg Cu/kg. They were then divided into three groups. One group continued to receive the control diet while another received a similar diet containing 560 mg Zn/kg. A third group, fed the control diet, received a subcutaneous dose of 90 mg Zn/kg body weight every 2-3 days for the duration of the experiment. Rats from each group were killed on days 7 and 14. Low Cu status in Zn-treated rats was indicated by lower than normal serum Cu concentration, serum ceruloplasmin activity, low liver and kidney Cu concentrations and low cytochrome C oxidase activity. None of these changes, however, were related to an increase in Cu as a result of Zn-induced MT in the intestinal epithelial cell. Instead, as the MT concentrations rose, Ca concentration decreased. This study suggests that the effects of high Zn treatment on Cu status are not the result of the longheld theory that Zn-induced intestinal MT sequesters Cu and prevents its passage to the circulation. Instead, it may be caused by a direct effect of high lumenal Zn concentrations on Cu transport into the epithelial cell. KEY WORDS: Zinc, Copper, MetaUothionein, Intestine, Kidney, Liver, Absorption, Rat
1Presented in part at the Federation of the American Society of Experimental Biology Meetings in Atlanta, Ga, 1991. 2Correspondence to this author. 3Present address: Sanofi Diagnostics Pasteur, Inc., Chaska, MN 55318 4Present address: Department of Animal and Nutritional Sciences, Kendall Hall, University of New Hampshire, Durham, NH 03824-3590.
897
898
P.G. REEVES et al. INTRODUCTION
Previous reports have shown that large doses of zinc (Zn) given to animals or humans, either by diet or by injections, depress (Cu) status. It has been shown that the major site for this interaction between Zn and Cu occurs in the intestine where Zn reduces the absorption of Cu (1-3). Many hypotheses have been presented to explain why Zn reduces Cu absorption. One proposes that Zn competes with Cu for a Cu-binding protein in the mucosal cell that is thought to aid Cu absorption, and reduces the potential for Cu to be absorbed (4). Another suggests that Cu binds tightly to Zn-induced metallothionein 0VIT) in the intestinal epithelial cells (5-7). This bound Cu, then, is not available for transport across the basolateral membrane into the circulation. In time, the MT-bound Cu is lost through the intestinal tract as epithelial cells slough off. The latter hypothesis is widely accepted as the mechanism by which large doses of Zn impair Cu absorption (8-10). It is also thought to be the mechanism by which Wilson's patients are detoxified of Cu when treated with Zn (11,12). If Zn-induced intestinal MT effectively binds Cu in vivo, it seems reasonable to conclude that the concentration of Cu would increase as the MT concentration increases. During recent studies we found that this was not the case. When rats were given high parenteral Zn, MT concentrations of intestinal epithelial cells increased 34-fold. However, concomitantly, the concentration of Cu increased less than 1.3-fold (13). Because these results did not confoma to the hypothesis that Zn-induced NIT effectively binds Cu in vivo, we did the following study to further examine these events.
METHODS
Sixty-three male Spmgue-Dawley rats (Sasco, Lincoln, NE) 5 weighing about 160 g were divided into three groups of 21 rats each. Rats in all three groups received a diet containing, by analysis, 42 mg Zn and 6 mg Cu/kg diet (Table 1) for one week. One group of rats continued to receive this diet for the duration of the experiment. Rats in a second group were fed a similar diet, but supplemented with additional Zn; it contained 560 mg of Zn/kg. A third group received the same diet as the first group and, in addition, each rat was given a subcutaneous injection of an emulsion of sunflower oil and Zn carbonate. The dose was given every 2-3 days in a portion equal to 1 mL of emulsion per kg body weight (BW) for the duration of the experiment. The dose was equivalent to 90 mg Zn/kg BW. This procedure was used to simulate continuous Zn absorption by allowing Zn to be slowly released into the circulation. As a result, rats fed the high-Zn diet consumed about 56 mg Znkg 9 : d * -z while those in the parenteral-Zn group received about 45 mg Znkg * l d 9 -1. After the rats had been on these regimens for one or two weeks, 7 and 14 rats, respectively, from each group were anesthetized with intraperitoneal injections of 50 mg sodium pentobarbital. Blood was drawn from the abdominal aorta into Monovette tubes (Sarstedt, Newton, NC) and allowed to clot at room temperature for one hour. Blood was centrifuged at 1,500g for 20 min and the serum was frozen until analyzed. A ten-centimeter section of the upper intestine, beginning at the pyloms, was excised and the lumen contents washed out with ice cold saline. The segment was slit open and the mucosal lining
SMention of a trademark or proprietary product does not constitute a guarantee or warranty of the product by the U.S. Department of Agriculture, and does not imply its approval to the exclusion of other products that may also be suitable.
METALLOTHIONEIN AND Cu METABOLISM
899
scraped off with the edge of a glass slide. Intestinal scrapings were frozen at -20 ~ C until analyzed. Kidneys and part of the liver from each rat were excised and flozen.
TABLE 1
Composition of the Basal Diet
Ingredients Dextrose 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Corn starch 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Casein, high protein 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Corn oil4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Soybean oil s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Egg white solids 6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . AIN-76 Mineral mix 7 Celluloses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . AIN-76 Vitamin mix 9 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Choline premix ~~ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Biotin premix H . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
g/gg 305 300 150 50 50 50 35 30 10 10 10
IlCN Biochemicals, Cleveland, OH, Cat. #901521. 2Argo, CPC International, Englewood Cliffs, NJ. 3Teklad, Madison, WI, Cat. #160030. 4Mazola Oil, Best Foods, CPC International, Inc., Englewood Cliffs, NJ. 5Crisco Oil, Procter & Gamble, Cincinnati, OH. ~l'eklad, Madison, WI, Cat. #160230. 7Teklad, Madison, WI, Cat. #170915. aTeklad, Madison, WI, Cat. #160390. 9Teklad, Madison, Wl, Cat. #40077.1~ g of choline bitartrate per kg in powdered sugar. H80 mg of d-biotin per kg in powdered sugar.
Proteins were precipitated from 0.5 mL of serum with 1.0 mL deionized water and 0.5 mL of 0.45 mol/L 5-sulfosalicylic acid in water. The mixture was allowed to sit at room temperature for one hr and then centrifuged at 2,000g for 20 min. The resulting supematant was analyzed for Zn and Cu by flame atomic absorption spectroscopy (AAS). Fresh serum was analyzed for cemloplasmin activity (CPA) by the method of Schosinsky et al. (14). Intestinal scrapings were divided into two portions. One portion was homogenized in five volumes of 0.23 mol/L 5-sulfosalicylic acid and allowed to stand at room temperature for one hr. After centfifugation, the supematant was diluted appropriately and analyzed for Zn and Cu by AAS. The remaining portion was analyzed for MT by the ~~ displacement method of Eaton and Toal (15). Portions of the kidneys and livers were dry-ashed in a muffle fumace. Residue from each sample was diluted to 5 mL in 0.1 mol/L HC1 and analyzed for Zn and Cu by AAS. Another portion of each tissue was homogenized in appropriate buffers and analyzed for the activities of Cu-Zn superoxide dismutase (SOD) (16) and cytochrome-C oxidase (CCO) (17). Still another portion of each tissue was analyzed for MT concentration. The data were analyzed by a two-way analysis of variance procedure (ANOVA; Crunch Statistical Software, Oakland, CA). Validity of the A_NOVA is predicated upon the assumption that the variances are homogeneous. A test for homogeneity was done for each ANOVA by the method of Hartley (18). When the test was significant, i.e., variances were not homogeneous, the data were transfomaed to achieve
900
P.G. REEVES et al.
or to approach homogeneity and analyzed. Data for each of the transformed parameters are presented in Table 2. Because the number of subjects per group varied greatly between weeks on experiment, Tukey's (19) methods were used to compare means. Although the analysis included both time and treatment, significant differences between means are shown only between treatments within each period.
TABLE 2 Ln Transform ation of Data ~
Week One of Experiment Control
Hi Diet Zn
Hi Par. Zn 2
3,89:t0.06
3.90-20.06
Intestinal Epithelial cells Zn 5.55s MT 1.84:k0.13
6.90-k0.10 4.52:t0.21
Liver Zn MT
6.20i-0.03 1.78-20.07
Kidney Zn MT
6.01i-0.02 1.93-&-0.11
Serum Zn
3.14:t0.03
Week Two of Experiment Control
Hi Diet Zn
Hi Par. Zn
2.99_+0.02
3.36:i'0.00
4.31:t0.05
6.73s 4.47s
5.75x'-0.03 1.53~0.13
6.83+0.06 3.60-s
7.35!-0.03 5.08:t0.06
6.34x'-0.05 2.07i-0.22
7.08i-0.07 3.96x'-0.12
6.21i-0.03 1.79i-0.03
6.22!~0.07 1.93:t0.05
7.16i/).04 4.37i-0.10
6.16_+0.00 2.33_+0.15
6.43x'-0.03 3.05-20.14
5.93i-0.02 1.81_+0.09
5.99x'-0.03 2.03-+0.09
6.80-1-0.00 4.13-+0.10
1When the test for homogeneity of variance between groups for each variable was significant by Hartley's F~,~ test, a natural log transformation of the data was performed. An ANOVA was then performed on this transformed data. 2High parenteral Zn. RESULTS
Throughoutthe two-week experiment, there were no significant effects of treatment on weight gain of rats (Means + SEM: control, 6.7 + 0.3; high dietary Zn, 6.8 + 0.2; high parenteral Zn, 6.9 + 0.2 g/d). Serum. Table 3 shows the effects of high Zn treatment on serum parameters. There were highly significant effects of treatment on serum Zn and Cu. During the first week, both high dietary Zn and high parenteral Zn caused an elevation in serum Zn to as much as 2.5 times that of the controls. However, after two weeks on the regimens, semm Zn in rats fed high dietary Zn had returned to near noma al. However, in rats treated with high parenteral Zn, serum Zn was even higher than the previous week (P<0.001). After one week, semm Cu concentrations were significantly (P<0.001) depressed in rats fed high dietary Zn but were significantly (P<0.001) elevated in those given parenteral Zn, as compared to control rats. After two weeks, however, semm Cu concentrations Were severely depressed (P<0.001) in both groups treated with high Zn when compared to controls. Serum CPA followed the same course as
METALLOTHIONEIN AND Cu METABOLISM
901
serum Cu. By the end of the second week, CPA in the group treated with high parenteral Zn was barely detectable in 10 of 14 rats.
TABLE 3
Effect of High Dietary or High Parenteral Zn on the Concentrations of Zn and Cu and on the Activity of Cemloplasmin (CPA) in Serum of Rats
Week One of Experiment 1
Zn, pmol/L Cu, Rmol/L CPA,/akat/L
Week Two of Experiment 2
Control
Hi Diet Zn
Hi Par. Zn 3
Control
Hi Diet Zn
Hi Par. Zn
23.31-0.7 a 13.1i-0.4 a 2.08-.L-0.08~
49.73+3.1 b 9.2+0.9 b 0.94:t0.25 b
56.6+7.0 b 20.2i-0.3 c 3.7 li,0.15 c
19.9x'-0.4 a 29.1:1:1.2b 75.4+3.9 ~ 14.5-+0.4 a 5.5-+0.4 b 4.6-+0.6 b 2.43=~-0.05a 0.42:1.-0.13b 0.1 li-0.08 b
'Mean + SEM of 7 (1st wk) and 14 (2nd wk) replicates per mean. Means across groups with different superscript letters are significantly different from each other at P<_0.05.2Means--I:SEMof 14 replicates per mean. 3High parenteral Zn.
Intestinal Epithelial Cells. The effects of Zn treatment on Zn, MT and Cu in the intestinal epithelial cells ale presented in Figure 1 and Table 2. After one week on the regimens, changes in Zn concentrations in the epithelial cells were similar to changes in MT concentrations (Figure 1, Panels Zn & MT). Thmugbout the study, MT and Zn concentrations were significantly (P<0.001) higher in the epithelial cells of rats treated with Zn than in contmls. However, at the end of the second week, MT concentrations in the epithelial cells of rats fed high Zn diets were significantly lower (P<0.001) than at the end of the first week. When we subtracted the basal levels of Zn and MT from the stimulated values, the molar ratio of Zn:MT for both periods ranged from 6.6:1 to 7.8:1 for all groups except those given high-Zn in the diet for two weeks; this value was 15:1. When we took into account only those rats treated with Zn and determined the relationship between MT and Zn concentrations in individual rats, we found a highly significant positive correlation (Figure 2). Treating rats with high Zn had a negative effect on the Cu concentration in the intestinal epithelial cells (Figure 1, Panel 3). Both Zn treatments depressed Cu in the intestinal epithelial ceils at both periods of sampling but was significant only after two weeks (P<0.01). Figure 2 shows no correlation between MT and Cu in the Zn-treated rats, when the data were expressed on the basis of individual animals. Liver. Figure 3 shows the effects of high Zn treatment on liver parameters. There was a two-fold elevation in liver Zn concentration in the groups given high parenteral Zn, as compared to those fed high dietary Zn or control diets. Liver Zn appeared to be slightly elevated by high dietary Zn after one week, but returned to control values by the end of the second week of treatment. Parenteral Zn but not high dietary Zn caused a large change in liver MT concentration (Figure 3, Panel MT). After one week of treatment, high parenteral Zn caused a six-fold increase in the concentration of M T over those receiving dietary Zn treatments. After two weeks, there was a twelve-fold difference. The effects of Zn treatment on liver Cu are shown in Figure 3, Panel C. The difference in
902
P.G. REEVES et al.
liver Cu concentrations among groups were not significant after the first week of treatment. However, after the second week, both groups treated with Zn were similar but lower t h ~ the controls.
TABLE 4 Effect of High Dietary or High Parenteral Zn on the Activities of Cytochrome-C Oxidase (CCO) and Cu, Zn-Supetoxide Dismutase (SOD) in Liver and Kidney of Rats
Week One of Experiment ~ Control
Hi Diet Zn
Week Two of Experiment 2
Hi Par. Zn 3
Control
Hi Diet Zn
Hi Par. Zn
Liver CCO 4 SOD
2.66i-0.30 271+8
2.33:t-0.16 258+15
2.37:t0.16 297+8
2.85i-0.21 a 264+6
2.12:t0.10 b 232+10
1.89:~.13 b 219+16
I~dney CCO SOD
2.60-!-0.19 243+ 14
2.59:!-0.14 253+4
2.14:t0.12 271 + 11
2.51_+0.12 239+8
2.48i-0.11 245+6
2.23!-0.08 231 +6
~Meard:SEM of 7 replicates per mean. Means across groups with different superscript letters are significantly different from each other a 1~__0.05. :Means.-I:SEMof 14 replicates per mean. 3High parenteral Zn. 4CCO and SOD activities are expressed as nkat/mg protein.
To deten~ine the severity of Cu status we measured the activities of two Cu-dependent enzymes in liver. Table 4 shows that both Zn treatments depressed CCO activity at both periods of measurement. However, the difference was significant (P<0.02) only after two weeks on the regimens. There was a slight but non-signillcant (P<0.06) effect on SOD activity during the second week in the group treated with high parenteral Zn compared to the control group. Kidney. Figure 4 shows the effect of Zn treatment on kidney parameters. After one week on the regimens, the concentration of Zn was significantly (P<0.001) increased in the kidneys of rats given high parenteral Zn as compared to the control rats. High dietary Zn also tended to elevate kidney Zn, but not significantly (P<0.06). After two weeks, kidney Zn in rats given high parenteral Zn had risen to more than two-fold higher than that in either the control or the high dietary Zn groups. MT concentration in kidneys followed the same pattern as Zn (Figure 4, Panel MT). There was no significant effect of high dietary Zn treatment on MT concentration in the kidneys compared to those fed the control diets. At the end of the first week of treatment, the MT concentration was two-fold higher in the group treated with parenteral Zn than in the other two groups. By the end of the second week, MT concentration in the kidneys of the rats receiving parenteral treatment had increased to about nine times that of the other two groups. The effect of Zn treatment on kidney Cu is shown in Figure 4, P~aael Cu. One week on either nigh Zn treatments tended to depress the concentration of kidney Cu compared to controls, but the difference was not significant. After two weeks, however, both Zn treatments had depressed the
METALLOTHIONEIN AND Cu METABOLISM concentration of kidney Cu and the difference was highly significant (P<0.001).
903
2000 T
1600 l Z n
I~
Treating rats with high dietary Zn did not have a significant effect on Cu-dependent enzyme activities in kidney (Table 4). However, high parenteral Zn tended to lower CCO activity at both periods of measurement. Kidney SOD activity was affected only after two weeks of treatment.
DISCUSSION
It has been Well documented that treatment of animals or humans with large doses of Zn compromises their Cu status (1-4,10). It is believed that this effect is caused by the inhibition of Cu absorption when Cu hinds to Zn-induced MT in the intestinal epithelial cells. This hypothesis is based on the demonstrations that MT has a greater affinity for Cu than for Zn and will preferentially bind Cu. It is reasonable to conclude, therefore, that Cu would displace Zn on Zn-induced MT in the intestinal epithelial cells and render it unavailable for transfer across the serosal membranes. It is also reasonable to conclude that the concentration of epithelial Cu would increase as MT concentration increased. Although this hypothesis has been widely accepted, data from past experiments, as well as the present one, do not support it.
20 15 10 5 0
Cu
1
2
Weeks on Experiment FIG 1. Effect of normal dietary Zn (open), high dietary Zn (diag.), and high parenterai Zn (closed) on the concentration of Zn, MT, and Cu in intestinal mucosa. Bars represent the mean + SEM of 7 (lst wk) and 14 (2nd wk) replicates. Bars within each panel with different superscripts are significantly different with P<0.05.
Three studies have been used to support the hypothesis that Zn-induced MT binds Cu in the intestinal epithelial cell and renders it unavailable for absorption. Ogiso et al. (7) were the first to suggest this mechanism when they found that 10 g Zn/kg diets reduced Cu absorption in rats. There was a greater proportion of Cu bound to Sephadex-isolated MT from these rats than from those not fed high Zn. However, they also found that the high-Zn diet reduced the concentration of Cu in the intestinal tract by 50% or more. Fischer et al. (5) detemained the effects of feeding various concentrations of dietary Zn on Cu transfer through the mucosal and serosal membranes of isolated intestinal loops of rats. Although them was a significant decline in Cu transport of rats fed between 30 and 240 mg Zn/kg diet, there was not a significant change in the amount of Cu in the intestinal epithelial cells. They did not measure MT concentration, but based on data from other studies it could be estimated that much more MT would have been present in the intestinal epithelial cells of rats fed the higher amount of Zn than in those fed the lower amount. They did, however, isolate MT by gel filtration and detemained the amount of Cu bound to the MT fraction. As dietary Zn increased eight-fold, Cu concentration in the MT fraction increased only 1.3-fold. They concluded that the mucosal Cu was bound almost exclusively to the protein fraction with a molecular weight similar to MT. They further concluded that Cu was bound to MT inside the
904
P.G. REEVES et al. mucosal cell ~ d thus inllibited from passing into the circulation.
2500 2000 1500 o D
._ I--.*
~
1000 500
" ~ ' OO - -
9
"
R2 = 0.71
P
0
-6 E
50 9
w
20 10 'q'v
vv
9
0 ....
-50
i ....
0
i ....
50
i ....
100
i ....
i ....
i ....
150 200 250 300
MT, /zmol/kg Tissue 2. Correlation between Zn and MT (closed circles) and between Cu and MT (closed triangles) concentrations in the intestinal mucosa. Values are for individual rats in all high-Zn treated groups, w e e k s 1 and 2 combined. Broken lines represent the 95% confidence internals. F I G
Work by a thin1 group, Hall et al. (20), also suggested that intestinal MT was the causative factor in Zn-induced reduction of Cu absorption. This group fed rats four concentrations of Zn ranging flom 30 to 900 mg/kg diet and found that ~*Cu absorption was depressed only in those rats fed the highest level of Zn. Total Cu concentration in the intestinal epithelial cells was not measured, but they found a three-fold increase in ~Cu in the intestinal mucosa, but only in the highest Zn group. MT was separated f-rom the mucosal supematant by gel filtration and the amount detemained by measuring the concentration of Zn in each fraction; the assumption being that the amount of MT was proportional to the amount of Zn. Mucosal contents of MT in this fraction rose 27-fold in the high-Zn group but the concentration of Cu only increased four-fold. However, they sumaised that flais small change in MT-bound Cu might interfere with the transfer of Cu across the basolateral membrane.
The basic assumption in all these reports was that the elevated concentration of Cu found in MT isolated by gel filtration was representative of the foma of Cu in the intestinal cell in vivo. This assumption may not be valid. Intact cells are highly compartmentalized, and under usual conditions, Cu and Zn ale tightly bound to specific enzymes at active sites, or less tightly bound to other proteins and small, diffusible llgands. It is possible that Cu is compartmentalized separate from MT, and when the compartments are broken apart by homogenization, as in all these procedures, Cu would be redistributed from ligands of weak affinity to those of stronger affinity, such as MT. Themetal composition of isolated MT may be very differem fiom that inside the cell (21). It cannot be assumed, therefore, that Cu is bound to Zn-induced MT in the cell just because it appears to be bound to isolated MT. The opposite state seems m ore typical. As a result of feeding high Zn, each of these studies either found a decrease in total cellular Cu or found no change in Cu concentration. Similar observations were made in the present study. That Cu status was greatly impaired by Zn treatment was clear flom the observations that serum Cu concentration and cerutoplasmin activity were depressed. In addition, liver and kidney Cu concentration and liver cytochmme C oxidase activity were significantly lower than in controls. However, none of these changes appeared to relate to the sequestering of Cu by Zn-induced MT in the intestinal epithelial cells. In fact, the opposite was true. MT concentration in epithelial cells of Zn-treated rats increased by as much as 30-fold over controls while the Cu concentration decreased by 30%. The reduction in Cu concentration inside the imestinal epithelial cell suggests that the major effect of high lumenal Zn might be the inhibition o f Cu transport into the cell, and from there into the circulation. Other work supports this suggestion. Oestreicher and Cousins (22) used a vascular-intestinal perfusion technique to study the effect of high lumenal Zn on Cu transport. In rats fed diets with Zn concentrations (30 mg/kg) too low to induce MT synthesis, they showed that elevated concentrations of lumenal Zn decreased Cu concentrations in the mucosal cytosol as well as the amount of Cu transported
METALLOTHIONEIN AND Cu METABOLISM to the vascular perfusate. Starcher (23), using chick s, showed that a single oral dose of 1 mg Zn per chick decreased the rate of 64Cu absorption by 67% and the amount of the isotope bound to epithelial proteins by 64%. Because these birds were not previously stimulated to produce abnormal MT concentrations in the intestinal cells, it suggests that Zn was inhibiting Cu absorption directly.
905
1600
I::Zn i i ,oo
Other direct support for our suggestion that Zn directly inhibits Cu uptake by intestinal epithelial cells without the influence of MT was presented by Van Campen (24). He fed rats a normal Zn diet (45 '01 mg/kg), a condition which does not elevate MT synthesis in the intestinal epithelial cells, and introduced, in situ, Zn and ~*Cu solutions into the Cu intestinal lumen so that the Zn:Cu ratio was approximately 150:1. After three hours, a 50% reduction in the body tissue uptake of 64Cu was 4o observed in rats receiving high concentrations of Zn. In previous experiments, Van Campen and Scharfe (2) 20 showed that Zn signillcantly reduced the amount of 0 - 64Cu leaving the intestinal lumen. The present study 1 2 suggests that during long-term treatment with high Weeks on Experiment dietary Zn, the intestinal epithelial cells may have adopted a mechanism other than MT synthesis to FIG 3. Effect of n o r m a l dietary Zn (open), high prevent excessive Zn absorption. After one week of dietary Zn (diag.), and high parenteral Zn (closed) consuming high-Zn diets, semm Zn concentrationwas on the concentration of Zn, MT, and Cu in liver. double that of the control group. Both mucosal Zn Bars represent the mean + SEM of 7 (lst wk) and and MT concentrations were elevated and the 14 (2nd wk) replicates. Bars within each panel calculated molar ratio of Zn to newly formed MT was with different superscripts are significantly 7.5:1. This is close to the maximal binding capacity different with P<0.05. of MT for Zn. However, after two weeks on this regimen, serum Zn was reduced by 40% compared to the first week, and although mucosal Zn remained unchanged, MT concentration was reduced by more than 60%. The Zn:MT molar ratio in this case was 15:1. This suggests that the mucosal cells had begun to adapt to high dietary Zn by lowering the synthesis of MT but retaining some other mechanism to prevent excessive Zn accumulation in the body.
9 oI
I 9 9
Oral Zn therapy is very effective in preventing liver damage caused by excessive Cu deposition in Wilson's disease patients (12), Using a rat model, Lee et al. (11) loaded rats with Cu (750 mg Cu/kg diet) for eight weeks. Depot Zn treatment, much like that in the present experiment, was begun at three weeks aud continued for eight more weeks. As a result of Zn treatment, SGPT activity was reduced, liver Cu was unchanged, liver MT increased five-fold, liver Zn increased about two-fold, and Cu in the liver cytosol (25,000g supemataut) decreased almost two-fold. They interpreted these results to indicate that Zn protected the liver from Cu toxicity by inducing MT which bound Cu. However, Zn-induced MT in the cytosol did not bind Cu. On the other hand, Zn treatment caused almost half of the cellular Cu to remain with the 25,000g pellet. This suggests that if MT binds Cu, much of it became associated with the particulate fraction of the cell. The question remaining, however, is what artifacts are introduced by the redistribution of Cu during homogenization and centrifugation? It has been shown that liver Cu in Cu-loaded animals (25,26) and humans (27) is associated with the particulate fraction. Much of it can
906
P.G. R E E V E S et al.
be extracted only with 2-mercaptoethanol while using freeze-thaw procedures.
1200 1000
Zn The situation in the present study was much different from that described by Lee et at. (11). It could be argued that the reduced concentration of Cu in liver and kidney, because of high Zn treatment, was because high-Zn prevented Cu absorption from the diet. Therefore, not enough Cu was available to maintain nomaal tissue concentrations.
800 6oo 4o0
o (n (/)
200 0 70
4-,
60
MT
!
50 40
.~ ~ E
30 20 10 140
Cu
(3
120 100 80 6O
40 20 0 1
2
Week s on E x p e r i m e n t
FIG 4. Effect of normal dietary Zn (open), high dietary Zn (diag.), and high parenteral Zn (closed) on the concentration of Zn, MT, and Cu in kidney. Bars represent the mean _+ SEM of 7 (lst wk) and 14 (2rid wk) replicates. Bars within each panel with different superscripts are significantly
In summary, it was clear from many observations that Cu status was impaired in rats receiving the high Zn treatments, high dietary Zn or high parenteral Zn. Semm Cu concentration and cemloplasmin activity were depressed, liver and kidney Cu concentrations and liver cytochrome-C oxidase activity were all significantly depressed. None of these changes, however, appeared to be related to the removal of Cu by Zn-induced MT in the intestinal epithelial cell. Instead, results showed that total intestinal mucosal Cu decreased when MT was induced by high Zn administration. This suggests that entry of Cu into the intestinal epithelial cells is inhibited by the presence of high Zn. This is the most likely mechanism by which high Zn impairs Cu status, and not because Zn-induced MT sequesters Cu and prevents its transfer outof the cell
ACKNOWLEDGEMENTS
The authors wish to acknowledge the technical assistance of Brenda Skinner for the enzymatic and MT analyses, Terry Shuler mid his staff for mineral analyses, Denice Schafer and her staff for care of the rats, and Made Swenson for manuscript preparation and management.
LITERATURE CITED
1.
Magee AC, Matrone G. Studies on growth, copper metabolism and iron metabolism of rats fed high levels of zinc. J Nutr 1960;72:233-42.
2.
Van Campen DR, Scaife PU. Zinc interference with copper absorption in rats. J Nutr 1967;91:473-6.
3.
Ogiso T, Mariyama K, Sasaki S, Ishimura Y, Minato A. Inhibitory effect of high dietary zinc on copper absorption in rats. Chem Phann Bull 1974;22:55-60.
METALLOTHIONEIN AND Cu METABOLISM
907
4.
Evans GW, Majors PF, Comatzer WE. Mechanism for cadmium and zinc antagonism of copper metabolism. Biochem Biophys Res Commun 1970;40:1142-8.
5.
Fischer PWF, Giroux A, L'Abb6 MR. The effect of dietary zinc on intestinal copper absorption. Am J Clin Nutr 1981;34:1670-5.
6.
Fischer PWF, Giroux A, L'Abbe MR. Effects of zinc on mucosal copper binding and on the kinetics of copper absorption. J Nutr 1983;113:462-9.
7.
Ogiso T, Ogawa N, Miura T. Inhibitory effect of high dietary zinc on copper absorption in rats. II. Binding of copper and zinc to cytosol proteins in the intestinal mucosa. Chem Pharm Bull 1979;27:515-21.
8.
Cousins RJ. Absorption, transport, and hepatic metabolism of copper ~aad zinc: special reference to metallothionein and cemloplasmin. Physiol Rev 1985;65:238-309.
9.
Harris ED. Copper transport: An overview. Pmc Soc Exp Biol Med 1991;196:130-40.
10.
O'Dell BL. Copper. In: Brown ML, ed. Present Knowledge in Nutrition, Washington, DC, International Life Sciences Institute, Nutrition Foundation, 1990:261-7.
11.
Lee D-Y, Brewer G J, Wang Y. Treatment of Wilson's disease with zinc. VII. Protection of the liver t o m copper toxicity by zinc-induced metallothionein in a rat model. J Lab Clin Meal 1989; 114:639-46.
12.
Brewer GJ, Yuzbasiyan-Gurkan V, Lee D-Y. Use of zinc-copper metabolic interactions in the treatment of Wilson's disease. J Am Coll. Nutr 1990;9:487-91.
13.
Reeves PG, Nelson KL. Effect of zinc pretreatment on cisplatin toxicity and trace element metabolism in rats. l Trace Elem Exp Med 1991;4:89-10I.
14.
Schosinsky KH, Lehmann HP, Beeler MF. Measurement of eemloplasmin from its oxidase activity in serum by use of o-dianisidine dihydrochloride. Clin Chem 1974;20: 1556-63.
15.
Eaton DL, "Foal BF. Evaluation of the Cd/hemoglobin affinity assay for the rapid detemaination of metallothionein in biological tissues. Toxicol Appl Pharmacol 1982;66: 134-42.
16.
Marklund S, Marklund G. Involvement of superoxide anion radical in the auto-oxidation of pyrogallol and a convenient assay for superoxide dismutase. Eur J Biochem 1974;47:469-74.
17.
Prohaska JR, Wells WW. Copper deficiency in the developing rat brain: A possible model for Menkes' steely-hair disease. J Neumchem 1974;23:91-8.
18.
Hartley HO. The maximum F-ratio as a short-cut test for heterogeneity of variance. Biometrika 1950;37:308-12.
19.
Tukey JW. Comparing individual means in the analysis of variance. Biometrics 1949;5:99-114.
20.
Hall AC, Young BW, Bremner I. Intestinal metallothionein and the mutual antagonism between copper and zinc in the rot. J Inorg Biochem 1979; 11:57-66.
908
P.G. REEVES et al.
21.
Bremner I. Involvement of metallothionein in the hepatic metabolism of copper. J Nutr 1987; 117:19-29.
22.
Oestreicher P, Cousins RJ. Copper and zinc absoiption in the rat: mechmaism of mutual antagonism. J Nutr 1985;115:159-66.
23.
Starcher BC. Studies on the mechanism of copper absorption in the chick. J Nutr 1969; 97:321-6.
24.
Van Cmapen DR. Effects of zinc, cadmium, silver and mercury on the absorption and distribution of copper-64 in rats. J Nutr 1966;88:125-30.
25.
Mehra RK, Bremner I. Species differences in the occurrence of copper-metallothionein in the particulate fractions of the liver of copper-loaded animals. Biochem J 1984;219: 539-46.
26.
Johnson GF, MoreU AG, Stockert RJ, Stemlieb I. Hepatic lysosomal copper protein in dogs with an inherited copper toxicosis. Hepatology 1981;1:243-8.
27.
Riordan JR, Richards V. Human fetal liver contains both zinc-and copper-rich folms of metallothionein. J Biol Chem 1980;255:5380-3.
Accepted f o r p u b l i c a t i o n
September 23, 1993.
Pergamon
0271-5317(94)E0023-W
Z I N C - I N D U C E D M E T A L L O T H I O N E I N AND C O P P E R M E T A B O L I S M IN INTESTINAL MUCOSA, LIVER, AND KIDNEY O F RATS 1 Philip G. Reeves 2, Kerry L. Rossow 3, and Dennis J. Bobilya 4
United States Department of Agriculture Agricultural Research Service Grand Forks Human Nutrition Research Center Grand Forks, ND 58202-9034 ABSTRACT
Large doses of parenteral zinc (Zn) and/or the feeding of high Zn diets to animals or humans for long periods affects copper (Cu) metabolism. Previous work suggests that Zninduced metallothionein (Mr) in intestinal epithelial ceils binds Cu and inhibits its absorption. This study was designed to determine the effects of treating rats with high dietary or high parenteral Zn on Cu metabolism and its relationship to MT in the intestinal epithelium, liver and kidney. Six-week-old male rats were fed for one week a control diet containing 42 nag Zn and 6 mg Cu/kg. They were then divided into three groups. One group continued to receive the control diet while another received a similar diet containing 560 mg Zn/kg. A third group, fed the control diet, received a subcutaneous dose of 90 mg Zn/kg body weight every 2-3 days for the duration of the experiment. Rats from each group were killed on days 7 and 14. Low Cu status in Zn-treated rats was indicated by lower than normal serum Cu concentration, serum ceruloplasmin activity, low liver and kidney Cu concentrations and low cytochrome C oxidase activity. None of these changes, however, were related to an increase in Cu as a result of Zn-induced MT in the intestinal epithelial cell. Instead, as the MT concentrations rose, Ca concentration decreased. This study suggests that the effects of high Zn treatment on Cu status are not the result of the longheld theory that Zn-induced intestinal MT sequesters Cu and prevents its passage to the circulation. Instead, it may be caused by a direct effect of high lumenal Zn concentrations on Cu transport into the epithelial cell. KEY WORDS: Zinc, Copper, MetaUothionein, Intestine, Kidney, Liver, Absorption, Rat
1Presented in part at the Federation of the American Society of Experimental Biology Meetings in Atlanta, Ga, 1991. 2Correspondence to this author. 3Present address: Sanofi Diagnostics Pasteur, Inc., Chaska, MN 55318 4Present address: Department of Animal and Nutritional Sciences, Kendall Hall, University of New Hampshire, Durham, NH 03824-3590.
897
898
P.G. REEVES et al. INTRODUCTION
Previous reports have shown that large doses of zinc (Zn) given to animals or humans, either by diet or by injections, depress (Cu) status. It has been shown that the major site for this interaction between Zn and Cu occurs in the intestine where Zn reduces the absorption of Cu (1-3). Many hypotheses have been presented to explain why Zn reduces Cu absorption. One proposes that Zn competes with Cu for a Cu-binding protein in the mucosal cell that is thought to aid Cu absorption, and reduces the potential for Cu to be absorbed (4). Another suggests that Cu binds tightly to Zn-induced metallothionein 0VIT) in the intestinal epithelial cells (5-7). This bound Cu, then, is not available for transport across the basolateral membrane into the circulation. In time, the MT-bound Cu is lost through the intestinal tract as epithelial cells slough off. The latter hypothesis is widely accepted as the mechanism by which large doses of Zn impair Cu absorption (8-10). It is also thought to be the mechanism by which Wilson's patients are detoxified of Cu when treated with Zn (11,12). If Zn-induced intestinal MT effectively binds Cu in vivo, it seems reasonable to conclude that the concentration of Cu would increase as the MT concentration increases. During recent studies we found that this was not the case. When rats were given high parenteral Zn, MT concentrations of intestinal epithelial cells increased 34-fold. However, concomitantly, the concentration of Cu increased less than 1.3-fold (13). Because these results did not confoma to the hypothesis that Zn-induced NIT effectively binds Cu in vivo, we did the following study to further examine these events.
METHODS
Sixty-three male Spmgue-Dawley rats (Sasco, Lincoln, NE) 5 weighing about 160 g were divided into three groups of 21 rats each. Rats in all three groups received a diet containing, by analysis, 42 mg Zn and 6 mg Cu/kg diet (Table 1) for one week. One group of rats continued to receive this diet for the duration of the experiment. Rats in a second group were fed a similar diet, but supplemented with additional Zn; it contained 560 mg of Zn/kg. A third group received the same diet as the first group and, in addition, each rat was given a subcutaneous injection of an emulsion of sunflower oil and Zn carbonate. The dose was given every 2-3 days in a portion equal to 1 mL of emulsion per kg body weight (BW) for the duration of the experiment. The dose was equivalent to 90 mg Zn/kg BW. This procedure was used to simulate continuous Zn absorption by allowing Zn to be slowly released into the circulation. As a result, rats fed the high-Zn diet consumed about 56 mg Znkg 9 : d * -z while those in the parenteral-Zn group received about 45 mg Znkg * l d 9 -1. After the rats had been on these regimens for one or two weeks, 7 and 14 rats, respectively, from each group were anesthetized with intraperitoneal injections of 50 mg sodium pentobarbital. Blood was drawn from the abdominal aorta into Monovette tubes (Sarstedt, Newton, NC) and allowed to clot at room temperature for one hour. Blood was centrifuged at 1,500g for 20 min and the serum was frozen until analyzed. A ten-centimeter section of the upper intestine, beginning at the pyloms, was excised and the lumen contents washed out with ice cold saline. The segment was slit open and the mucosal lining
SMention of a trademark or proprietary product does not constitute a guarantee or warranty of the product by the U.S. Department of Agriculture, and does not imply its approval to the exclusion of other products that may also be suitable.
METALLOTHIONEIN AND Cu METABOLISM
899
scraped off with the edge of a glass slide. Intestinal scrapings were frozen at -20 ~ C until analyzed. Kidneys and part of the liver from each rat were excised and flozen.
TABLE 1
Composition of the Basal Diet
Ingredients Dextrose 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Corn starch 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Casein, high protein 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Corn oil4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Soybean oil s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Egg white solids 6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . AIN-76 Mineral mix 7 Celluloses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . AIN-76 Vitamin mix 9 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Choline premix ~~ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Biotin premix H . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
g/gg 305 300 150 50 50 50 35 30 10 10 10
IlCN Biochemicals, Cleveland, OH, Cat. #901521. 2Argo, CPC International, Englewood Cliffs, NJ. 3Teklad, Madison, WI, Cat. #160030. 4Mazola Oil, Best Foods, CPC International, Inc., Englewood Cliffs, NJ. 5Crisco Oil, Procter & Gamble, Cincinnati, OH. ~l'eklad, Madison, WI, Cat. #160230. 7Teklad, Madison, WI, Cat. #170915. aTeklad, Madison, WI, Cat. #160390. 9Teklad, Madison, Wl, Cat. #40077.1~ g of choline bitartrate per kg in powdered sugar. H80 mg of d-biotin per kg in powdered sugar.
Proteins were precipitated from 0.5 mL of serum with 1.0 mL deionized water and 0.5 mL of 0.45 mol/L 5-sulfosalicylic acid in water. The mixture was allowed to sit at room temperature for one hr and then centrifuged at 2,000g for 20 min. The resulting supematant was analyzed for Zn and Cu by flame atomic absorption spectroscopy (AAS). Fresh serum was analyzed for cemloplasmin activity (CPA) by the method of Schosinsky et al. (14). Intestinal scrapings were divided into two portions. One portion was homogenized in five volumes of 0.23 mol/L 5-sulfosalicylic acid and allowed to stand at room temperature for one hr. After centfifugation, the supematant was diluted appropriately and analyzed for Zn and Cu by AAS. The remaining portion was analyzed for MT by the ~~ displacement method of Eaton and Toal (15). Portions of the kidneys and livers were dry-ashed in a muffle fumace. Residue from each sample was diluted to 5 mL in 0.1 mol/L HC1 and analyzed for Zn and Cu by AAS. Another portion of each tissue was homogenized in appropriate buffers and analyzed for the activities of Cu-Zn superoxide dismutase (SOD) (16) and cytochrome-C oxidase (CCO) (17). Still another portion of each tissue was analyzed for MT concentration. The data were analyzed by a two-way analysis of variance procedure (ANOVA; Crunch Statistical Software, Oakland, CA). Validity of the A_NOVA is predicated upon the assumption that the variances are homogeneous. A test for homogeneity was done for each ANOVA by the method of Hartley (18). When the test was significant, i.e., variances were not homogeneous, the data were transfomaed to achieve
900
P.G. REEVES et al.
or to approach homogeneity and analyzed. Data for each of the transformed parameters are presented in Table 2. Because the number of subjects per group varied greatly between weeks on experiment, Tukey's (19) methods were used to compare means. Although the analysis included both time and treatment, significant differences between means are shown only between treatments within each period.
TABLE 2 Ln Transform ation of Data ~
Week One of Experiment Control
Hi Diet Zn
Hi Par. Zn 2
3,89:t0.06
3.90-20.06
Intestinal Epithelial cells Zn 5.55s MT 1.84:k0.13
6.90-k0.10 4.52:t0.21
Liver Zn MT
6.20i-0.03 1.78-20.07
Kidney Zn MT
6.01i-0.02 1.93-&-0.11
Serum Zn
3.14:t0.03
Week Two of Experiment Control
Hi Diet Zn
Hi Par. Zn
2.99_+0.02
3.36:i'0.00
4.31:t0.05
6.73s 4.47s
5.75x'-0.03 1.53~0.13
6.83+0.06 3.60-s
7.35!-0.03 5.08:t0.06
6.34x'-0.05 2.07i-0.22
7.08i-0.07 3.96x'-0.12
6.21i-0.03 1.79i-0.03
6.22!~0.07 1.93:t0.05
7.16i/).04 4.37i-0.10
6.16_+0.00 2.33_+0.15
6.43x'-0.03 3.05-20.14
5.93i-0.02 1.81_+0.09
5.99x'-0.03 2.03-+0.09
6.80-1-0.00 4.13-+0.10
1When the test for homogeneity of variance between groups for each variable was significant by Hartley's F~,~ test, a natural log transformation of the data was performed. An ANOVA was then performed on this transformed data. 2High parenteral Zn. RESULTS
Throughoutthe two-week experiment, there were no significant effects of treatment on weight gain of rats (Means + SEM: control, 6.7 + 0.3; high dietary Zn, 6.8 + 0.2; high parenteral Zn, 6.9 + 0.2 g/d). Serum. Table 3 shows the effects of high Zn treatment on serum parameters. There were highly significant effects of treatment on serum Zn and Cu. During the first week, both high dietary Zn and high parenteral Zn caused an elevation in serum Zn to as much as 2.5 times that of the controls. However, after two weeks on the regimens, semm Zn in rats fed high dietary Zn had returned to near noma al. However, in rats treated with high parenteral Zn, serum Zn was even higher than the previous week (P<0.001). After one week, semm Cu concentrations were significantly (P<0.001) depressed in rats fed high dietary Zn but were significantly (P<0.001) elevated in those given parenteral Zn, as compared to control rats. After two weeks, however, semm Cu concentrations Were severely depressed (P<0.001) in both groups treated with high Zn when compared to controls. Serum CPA followed the same course as
METALLOTHIONEIN AND Cu METABOLISM
901
serum Cu. By the end of the second week, CPA in the group treated with high parenteral Zn was barely detectable in 10 of 14 rats.
TABLE 3
Effect of High Dietary or High Parenteral Zn on the Concentrations of Zn and Cu and on the Activity of Cemloplasmin (CPA) in Serum of Rats
Week One of Experiment 1
Zn, pmol/L Cu, Rmol/L CPA,/akat/L
Week Two of Experiment 2
Control
Hi Diet Zn
Hi Par. Zn 3
Control
Hi Diet Zn
Hi Par. Zn
23.31-0.7 a 13.1i-0.4 a 2.08-.L-0.08~
49.73+3.1 b 9.2+0.9 b 0.94:t0.25 b
56.6+7.0 b 20.2i-0.3 c 3.7 li,0.15 c
19.9x'-0.4 a 29.1:1:1.2b 75.4+3.9 ~ 14.5-+0.4 a 5.5-+0.4 b 4.6-+0.6 b 2.43=~-0.05a 0.42:1.-0.13b 0.1 li-0.08 b
'Mean + SEM of 7 (1st wk) and 14 (2nd wk) replicates per mean. Means across groups with different superscript letters are significantly different from each other at P<_0.05.2Means--I:SEMof 14 replicates per mean. 3High parenteral Zn.
Intestinal Epithelial Cells. The effects of Zn treatment on Zn, MT and Cu in the intestinal epithelial cells ale presented in Figure 1 and Table 2. After one week on the regimens, changes in Zn concentrations in the epithelial cells were similar to changes in MT concentrations (Figure 1, Panels Zn & MT). Thmugbout the study, MT and Zn concentrations were significantly (P<0.001) higher in the epithelial cells of rats treated with Zn than in contmls. However, at the end of the second week, MT concentrations in the epithelial cells of rats fed high Zn diets were significantly lower (P<0.001) than at the end of the first week. When we subtracted the basal levels of Zn and MT from the stimulated values, the molar ratio of Zn:MT for both periods ranged from 6.6:1 to 7.8:1 for all groups except those given high-Zn in the diet for two weeks; this value was 15:1. When we took into account only those rats treated with Zn and determined the relationship between MT and Zn concentrations in individual rats, we found a highly significant positive correlation (Figure 2). Treating rats with high Zn had a negative effect on the Cu concentration in the intestinal epithelial cells (Figure 1, Panel 3). Both Zn treatments depressed Cu in the intestinal epithelial ceils at both periods of sampling but was significant only after two weeks (P<0.01). Figure 2 shows no correlation between MT and Cu in the Zn-treated rats, when the data were expressed on the basis of individual animals. Liver. Figure 3 shows the effects of high Zn treatment on liver parameters. There was a two-fold elevation in liver Zn concentration in the groups given high parenteral Zn, as compared to those fed high dietary Zn or control diets. Liver Zn appeared to be slightly elevated by high dietary Zn after one week, but returned to control values by the end of the second week of treatment. Parenteral Zn but not high dietary Zn caused a large change in liver MT concentration (Figure 3, Panel MT). After one week of treatment, high parenteral Zn caused a six-fold increase in the concentration of M T over those receiving dietary Zn treatments. After two weeks, there was a twelve-fold difference. The effects of Zn treatment on liver Cu are shown in Figure 3, Panel C. The difference in
902
P.G. REEVES et al.
liver Cu concentrations among groups were not significant after the first week of treatment. However, after the second week, both groups treated with Zn were similar but lower t h ~ the controls.
TABLE 4 Effect of High Dietary or High Parenteral Zn on the Activities of Cytochrome-C Oxidase (CCO) and Cu, Zn-Supetoxide Dismutase (SOD) in Liver and Kidney of Rats
Week One of Experiment ~ Control
Hi Diet Zn
Week Two of Experiment 2
Hi Par. Zn 3
Control
Hi Diet Zn
Hi Par. Zn
Liver CCO 4 SOD
2.66i-0.30 271+8
2.33:t-0.16 258+15
2.37:t0.16 297+8
2.85i-0.21 a 264+6
2.12:t0.10 b 232+10
1.89:~.13 b 219+16
I~dney CCO SOD
2.60-!-0.19 243+ 14
2.59:!-0.14 253+4
2.14:t0.12 271 + 11
2.51_+0.12 239+8
2.48i-0.11 245+6
2.23!-0.08 231 +6
~Meard:SEM of 7 replicates per mean. Means across groups with different superscript letters are significantly different from each other a 1~__0.05. :Means.-I:SEMof 14 replicates per mean. 3High parenteral Zn. 4CCO and SOD activities are expressed as nkat/mg protein.
To deten~ine the severity of Cu status we measured the activities of two Cu-dependent enzymes in liver. Table 4 shows that both Zn treatments depressed CCO activity at both periods of measurement. However, the difference was significant (P<0.02) only after two weeks on the regimens. There was a slight but non-signillcant (P<0.06) effect on SOD activity during the second week in the group treated with high parenteral Zn compared to the control group. Kidney. Figure 4 shows the effect of Zn treatment on kidney parameters. After one week on the regimens, the concentration of Zn was significantly (P<0.001) increased in the kidneys of rats given high parenteral Zn as compared to the control rats. High dietary Zn also tended to elevate kidney Zn, but not significantly (P<0.06). After two weeks, kidney Zn in rats given high parenteral Zn had risen to more than two-fold higher than that in either the control or the high dietary Zn groups. MT concentration in kidneys followed the same pattern as Zn (Figure 4, Panel MT). There was no significant effect of high dietary Zn treatment on MT concentration in the kidneys compared to those fed the control diets. At the end of the first week of treatment, the MT concentration was two-fold higher in the group treated with parenteral Zn than in the other two groups. By the end of the second week, MT concentration in the kidneys of the rats receiving parenteral treatment had increased to about nine times that of the other two groups. The effect of Zn treatment on kidney Cu is shown in Figure 4, P~aael Cu. One week on either nigh Zn treatments tended to depress the concentration of kidney Cu compared to controls, but the difference was not significant. After two weeks, however, both Zn treatments had depressed the
METALLOTHIONEIN AND Cu METABOLISM concentration of kidney Cu and the difference was highly significant (P<0.001).
903
2000 T
1600 l Z n
I~
Treating rats with high dietary Zn did not have a significant effect on Cu-dependent enzyme activities in kidney (Table 4). However, high parenteral Zn tended to lower CCO activity at both periods of measurement. Kidney SOD activity was affected only after two weeks of treatment.
DISCUSSION
It has been Well documented that treatment of animals or humans with large doses of Zn compromises their Cu status (1-4,10). It is believed that this effect is caused by the inhibition of Cu absorption when Cu hinds to Zn-induced MT in the intestinal epithelial cells. This hypothesis is based on the demonstrations that MT has a greater affinity for Cu than for Zn and will preferentially bind Cu. It is reasonable to conclude, therefore, that Cu would displace Zn on Zn-induced MT in the intestinal epithelial cells and render it unavailable for transfer across the serosal membranes. It is also reasonable to conclude that the concentration of epithelial Cu would increase as MT concentration increased. Although this hypothesis has been widely accepted, data from past experiments, as well as the present one, do not support it.
20 15 10 5 0
Cu
1
2
Weeks on Experiment FIG 1. Effect of normal dietary Zn (open), high dietary Zn (diag.), and high parenterai Zn (closed) on the concentration of Zn, MT, and Cu in intestinal mucosa. Bars represent the mean + SEM of 7 (lst wk) and 14 (2nd wk) replicates. Bars within each panel with different superscripts are significantly different with P<0.05.
Three studies have been used to support the hypothesis that Zn-induced MT binds Cu in the intestinal epithelial cell and renders it unavailable for absorption. Ogiso et al. (7) were the first to suggest this mechanism when they found that 10 g Zn/kg diets reduced Cu absorption in rats. There was a greater proportion of Cu bound to Sephadex-isolated MT from these rats than from those not fed high Zn. However, they also found that the high-Zn diet reduced the concentration of Cu in the intestinal tract by 50% or more. Fischer et al. (5) detemained the effects of feeding various concentrations of dietary Zn on Cu transfer through the mucosal and serosal membranes of isolated intestinal loops of rats. Although them was a significant decline in Cu transport of rats fed between 30 and 240 mg Zn/kg diet, there was not a significant change in the amount of Cu in the intestinal epithelial cells. They did not measure MT concentration, but based on data from other studies it could be estimated that much more MT would have been present in the intestinal epithelial cells of rats fed the higher amount of Zn than in those fed the lower amount. They did, however, isolate MT by gel filtration and detemained the amount of Cu bound to the MT fraction. As dietary Zn increased eight-fold, Cu concentration in the MT fraction increased only 1.3-fold. They concluded that the mucosal Cu was bound almost exclusively to the protein fraction with a molecular weight similar to MT. They further concluded that Cu was bound to MT inside the
904
P.G. REEVES et al. mucosal cell ~ d thus inllibited from passing into the circulation.
2500 2000 1500 o D
._ I--.*
~
1000 500
" ~ ' OO - -
9
"
R2 = 0.71
P
0
-6 E
50 9
w
20 10 'q'v
vv
9
0 ....
-50
i ....
0
i ....
50
i ....
100
i ....
i ....
i ....
150 200 250 300
MT, /zmol/kg Tissue 2. Correlation between Zn and MT (closed circles) and between Cu and MT (closed triangles) concentrations in the intestinal mucosa. Values are for individual rats in all high-Zn treated groups, w e e k s 1 and 2 combined. Broken lines represent the 95% confidence internals. F I G
Work by a thin1 group, Hall et al. (20), also suggested that intestinal MT was the causative factor in Zn-induced reduction of Cu absorption. This group fed rats four concentrations of Zn ranging flom 30 to 900 mg/kg diet and found that ~*Cu absorption was depressed only in those rats fed the highest level of Zn. Total Cu concentration in the intestinal epithelial cells was not measured, but they found a three-fold increase in ~Cu in the intestinal mucosa, but only in the highest Zn group. MT was separated f-rom the mucosal supematant by gel filtration and the amount detemained by measuring the concentration of Zn in each fraction; the assumption being that the amount of MT was proportional to the amount of Zn. Mucosal contents of MT in this fraction rose 27-fold in the high-Zn group but the concentration of Cu only increased four-fold. However, they sumaised that flais small change in MT-bound Cu might interfere with the transfer of Cu across the basolateral membrane.
The basic assumption in all these reports was that the elevated concentration of Cu found in MT isolated by gel filtration was representative of the foma of Cu in the intestinal cell in vivo. This assumption may not be valid. Intact cells are highly compartmentalized, and under usual conditions, Cu and Zn ale tightly bound to specific enzymes at active sites, or less tightly bound to other proteins and small, diffusible llgands. It is possible that Cu is compartmentalized separate from MT, and when the compartments are broken apart by homogenization, as in all these procedures, Cu would be redistributed from ligands of weak affinity to those of stronger affinity, such as MT. Themetal composition of isolated MT may be very differem fiom that inside the cell (21). It cannot be assumed, therefore, that Cu is bound to Zn-induced MT in the cell just because it appears to be bound to isolated MT. The opposite state seems m ore typical. As a result of feeding high Zn, each of these studies either found a decrease in total cellular Cu or found no change in Cu concentration. Similar observations were made in the present study. That Cu status was greatly impaired by Zn treatment was clear flom the observations that serum Cu concentration and cerutoplasmin activity were depressed. In addition, liver and kidney Cu concentration and liver cytochmme C oxidase activity were significantly lower than in controls. However, none of these changes appeared to relate to the sequestering of Cu by Zn-induced MT in the intestinal epithelial cells. In fact, the opposite was true. MT concentration in epithelial cells of Zn-treated rats increased by as much as 30-fold over controls while the Cu concentration decreased by 30%. The reduction in Cu concentration inside the imestinal epithelial cell suggests that the major effect of high lumenal Zn might be the inhibition o f Cu transport into the cell, and from there into the circulation. Other work supports this suggestion. Oestreicher and Cousins (22) used a vascular-intestinal perfusion technique to study the effect of high lumenal Zn on Cu transport. In rats fed diets with Zn concentrations (30 mg/kg) too low to induce MT synthesis, they showed that elevated concentrations of lumenal Zn decreased Cu concentrations in the mucosal cytosol as well as the amount of Cu transported
METALLOTHIONEIN AND Cu METABOLISM to the vascular perfusate. Starcher (23), using chick s, showed that a single oral dose of 1 mg Zn per chick decreased the rate of 64Cu absorption by 67% and the amount of the isotope bound to epithelial proteins by 64%. Because these birds were not previously stimulated to produce abnormal MT concentrations in the intestinal cells, it suggests that Zn was inhibiting Cu absorption directly.
905
1600
I::Zn i i ,oo
Other direct support for our suggestion that Zn directly inhibits Cu uptake by intestinal epithelial cells without the influence of MT was presented by Van Campen (24). He fed rats a normal Zn diet (45 '01 mg/kg), a condition which does not elevate MT synthesis in the intestinal epithelial cells, and introduced, in situ, Zn and ~*Cu solutions into the Cu intestinal lumen so that the Zn:Cu ratio was approximately 150:1. After three hours, a 50% reduction in the body tissue uptake of 64Cu was 4o observed in rats receiving high concentrations of Zn. In previous experiments, Van Campen and Scharfe (2) 20 showed that Zn signillcantly reduced the amount of 0 - 64Cu leaving the intestinal lumen. The present study 1 2 suggests that during long-term treatment with high Weeks on Experiment dietary Zn, the intestinal epithelial cells may have adopted a mechanism other than MT synthesis to FIG 3. Effect of n o r m a l dietary Zn (open), high prevent excessive Zn absorption. After one week of dietary Zn (diag.), and high parenteral Zn (closed) consuming high-Zn diets, semm Zn concentrationwas on the concentration of Zn, MT, and Cu in liver. double that of the control group. Both mucosal Zn Bars represent the mean + SEM of 7 (lst wk) and and MT concentrations were elevated and the 14 (2nd wk) replicates. Bars within each panel calculated molar ratio of Zn to newly formed MT was with different superscripts are significantly 7.5:1. This is close to the maximal binding capacity different with P<0.05. of MT for Zn. However, after two weeks on this regimen, serum Zn was reduced by 40% compared to the first week, and although mucosal Zn remained unchanged, MT concentration was reduced by more than 60%. The Zn:MT molar ratio in this case was 15:1. This suggests that the mucosal cells had begun to adapt to high dietary Zn by lowering the synthesis of MT but retaining some other mechanism to prevent excessive Zn accumulation in the body.
9 oI
I 9 9
Oral Zn therapy is very effective in preventing liver damage caused by excessive Cu deposition in Wilson's disease patients (12), Using a rat model, Lee et al. (11) loaded rats with Cu (750 mg Cu/kg diet) for eight weeks. Depot Zn treatment, much like that in the present experiment, was begun at three weeks aud continued for eight more weeks. As a result of Zn treatment, SGPT activity was reduced, liver Cu was unchanged, liver MT increased five-fold, liver Zn increased about two-fold, and Cu in the liver cytosol (25,000g supemataut) decreased almost two-fold. They interpreted these results to indicate that Zn protected the liver from Cu toxicity by inducing MT which bound Cu. However, Zn-induced MT in the cytosol did not bind Cu. On the other hand, Zn treatment caused almost half of the cellular Cu to remain with the 25,000g pellet. This suggests that if MT binds Cu, much of it became associated with the particulate fraction of the cell. The question remaining, however, is what artifacts are introduced by the redistribution of Cu during homogenization and centrifugation? It has been shown that liver Cu in Cu-loaded animals (25,26) and humans (27) is associated with the particulate fraction. Much of it can
906
P.G. R E E V E S et al.
be extracted only with 2-mercaptoethanol while using freeze-thaw procedures.
1200 1000
Zn The situation in the present study was much different from that described by Lee et at. (11). It could be argued that the reduced concentration of Cu in liver and kidney, because of high Zn treatment, was because high-Zn prevented Cu absorption from the diet. Therefore, not enough Cu was available to maintain nomaal tissue concentrations.
800 6oo 4o0
o (n (/)
200 0 70
4-,
60
MT
!
50 40
.~ ~ E
30 20 10 140
Cu
(3
120 100 80 6O
40 20 0 1
2
Week s on E x p e r i m e n t
FIG 4. Effect of normal dietary Zn (open), high dietary Zn (diag.), and high parenteral Zn (closed) on the concentration of Zn, MT, and Cu in kidney. Bars represent the mean _+ SEM of 7 (lst wk) and 14 (2rid wk) replicates. Bars within each panel with different superscripts are significantly
In summary, it was clear from many observations that Cu status was impaired in rats receiving the high Zn treatments, high dietary Zn or high parenteral Zn. Semm Cu concentration and cemloplasmin activity were depressed, liver and kidney Cu concentrations and liver cytochrome-C oxidase activity were all significantly depressed. None of these changes, however, appeared to be related to the removal of Cu by Zn-induced MT in the intestinal epithelial cell. Instead, results showed that total intestinal mucosal Cu decreased when MT was induced by high Zn administration. This suggests that entry of Cu into the intestinal epithelial cells is inhibited by the presence of high Zn. This is the most likely mechanism by which high Zn impairs Cu status, and not because Zn-induced MT sequesters Cu and prevents its transfer outof the cell
ACKNOWLEDGEMENTS
The authors wish to acknowledge the technical assistance of Brenda Skinner for the enzymatic and MT analyses, Terry Shuler mid his staff for mineral analyses, Denice Schafer and her staff for care of the rats, and Made Swenson for manuscript preparation and management.
LITERATURE CITED
1.
Magee AC, Matrone G. Studies on growth, copper metabolism and iron metabolism of rats fed high levels of zinc. J Nutr 1960;72:233-42.
2.
Van Campen DR, Scaife PU. Zinc interference with copper absorption in rats. J Nutr 1967;91:473-6.
3.
Ogiso T, Mariyama K, Sasaki S, Ishimura Y, Minato A. Inhibitory effect of high dietary zinc on copper absorption in rats. Chem Phann Bull 1974;22:55-60.
METALLOTHIONEIN AND Cu METABOLISM
907
4.
Evans GW, Majors PF, Comatzer WE. Mechanism for cadmium and zinc antagonism of copper metabolism. Biochem Biophys Res Commun 1970;40:1142-8.
5.
Fischer PWF, Giroux A, L'Abb6 MR. The effect of dietary zinc on intestinal copper absorption. Am J Clin Nutr 1981;34:1670-5.
6.
Fischer PWF, Giroux A, L'Abbe MR. Effects of zinc on mucosal copper binding and on the kinetics of copper absorption. J Nutr 1983;113:462-9.
7.
Ogiso T, Ogawa N, Miura T. Inhibitory effect of high dietary zinc on copper absorption in rats. II. Binding of copper and zinc to cytosol proteins in the intestinal mucosa. Chem Pharm Bull 1979;27:515-21.
8.
Cousins RJ. Absorption, transport, and hepatic metabolism of copper ~aad zinc: special reference to metallothionein and cemloplasmin. Physiol Rev 1985;65:238-309.
9.
Harris ED. Copper transport: An overview. Pmc Soc Exp Biol Med 1991;196:130-40.
10.
O'Dell BL. Copper. In: Brown ML, ed. Present Knowledge in Nutrition, Washington, DC, International Life Sciences Institute, Nutrition Foundation, 1990:261-7.
11.
Lee D-Y, Brewer G J, Wang Y. Treatment of Wilson's disease with zinc. VII. Protection of the liver t o m copper toxicity by zinc-induced metallothionein in a rat model. J Lab Clin Meal 1989; 114:639-46.
12.
Brewer GJ, Yuzbasiyan-Gurkan V, Lee D-Y. Use of zinc-copper metabolic interactions in the treatment of Wilson's disease. J Am Coll. Nutr 1990;9:487-91.
13.
Reeves PG, Nelson KL. Effect of zinc pretreatment on cisplatin toxicity and trace element metabolism in rats. l Trace Elem Exp Med 1991;4:89-10I.
14.
Schosinsky KH, Lehmann HP, Beeler MF. Measurement of eemloplasmin from its oxidase activity in serum by use of o-dianisidine dihydrochloride. Clin Chem 1974;20: 1556-63.
15.
Eaton DL, "Foal BF. Evaluation of the Cd/hemoglobin affinity assay for the rapid detemaination of metallothionein in biological tissues. Toxicol Appl Pharmacol 1982;66: 134-42.
16.
Marklund S, Marklund G. Involvement of superoxide anion radical in the auto-oxidation of pyrogallol and a convenient assay for superoxide dismutase. Eur J Biochem 1974;47:469-74.
17.
Prohaska JR, Wells WW. Copper deficiency in the developing rat brain: A possible model for Menkes' steely-hair disease. J Neumchem 1974;23:91-8.
18.
Hartley HO. The maximum F-ratio as a short-cut test for heterogeneity of variance. Biometrika 1950;37:308-12.
19.
Tukey JW. Comparing individual means in the analysis of variance. Biometrics 1949;5:99-114.
20.
Hall AC, Young BW, Bremner I. Intestinal metallothionein and the mutual antagonism between copper and zinc in the rot. J Inorg Biochem 1979; 11:57-66.
908
P.G. REEVES et al.
21.
Bremner I. Involvement of metallothionein in the hepatic metabolism of copper. J Nutr 1987; 117:19-29.
22.
Oestreicher P, Cousins RJ. Copper and zinc absoiption in the rat: mechmaism of mutual antagonism. J Nutr 1985;115:159-66.
23.
Starcher BC. Studies on the mechanism of copper absorption in the chick. J Nutr 1969; 97:321-6.
24.
Van Cmapen DR. Effects of zinc, cadmium, silver and mercury on the absorption and distribution of copper-64 in rats. J Nutr 1966;88:125-30.
25.
Mehra RK, Bremner I. Species differences in the occurrence of copper-metallothionein in the particulate fractions of the liver of copper-loaded animals. Biochem J 1984;219: 539-46.
26.
Johnson GF, MoreU AG, Stockert RJ, Stemlieb I. Hepatic lysosomal copper protein in dogs with an inherited copper toxicosis. Hepatology 1981;1:243-8.
27.
Riordan JR, Richards V. Human fetal liver contains both zinc-and copper-rich folms of metallothionein. J Biol Chem 1980;255:5380-3.
Accepted f o r p u b l i c a t i o n
September 23, 1993.