Zn is an essential mineral that acts as a cofactor for numerous enzymes and transcription factors(Reference McCall, Huang and Fierke1). The physiological responses to Zn deficiency are well-characterised in mammals(Reference Giugliano and Millward2–Reference Sandstead, Frederickson and Penland5), whereas less information is available regarding the effects of excess Zn intake. The pancreas has been suggested as the tissue most sensitive to excess Zn(Reference Sutomo, Woutersen and Van den Hamer6). The National Research Council (NRC)(7) proposes that Zn accumulates in tissues such as the liver, kidney and bone, in order to protect other organs from failure induced by Zn accumulation. Zn concentrations were increased in the liver (6-fold) and in the kidney (11-fold) in preruminant calves fed a diet supplemented with 500–700 mg Zn/kg, whereas the increases in Zn concentration in the heart and muscle were relatively smaller(Reference Jenkins and Hidiroglou8); similar results were obtained in sheep fed a diet supplemented with 700–2100 mg Zn/kg(Reference Henry, Littell and Ammerman9). Furthermore, Zn concentrations in the liver, kidney and bone were higher in rats fed a diet supplemented with 2438 mg Zn/kg compared with growing rats fed a diet containing 38 mg Zn/kg, and the Zn concentration of these tissues plateaued at supplementation levels of 2438–7238 mg Zn/kg(Reference Ansari, Miller and Neathery10).
Elevated Zn intake (3000 mg/kg) for a period of 14 d surprisingly induces growth in weaning piglets(Reference Hill, Mahan and Carter11–Reference Li, Yin and Li13). Considering that the Zn requirement for growing pigs is 100 mg/kg(14), the Zn-induced growth promotion results from the pharmacological effects of excess Zn intake. The physiological basis for these pharmacological effects remains unclear. ZnO possesses antimicrobial properties(Reference Hill, Mahan and Carter15), but several studies suggest that ZnO promotes growth in early-weaned and conventionally weaned pigs, regardless of diarrhoea prevalence or intestinal microbial numbers(Reference Poulsen16–Reference Katouli, Melin and Jensen-Waern18).
Zn homeostasis is primarily maintained by regulation of its absorption and secretion. Several Zn transporters of the Slc39 (Zip) and Slc30 (ZnT) families have been identified. Members of the Zip family have been shown to increase the cytosolic Zn concentration, whereas those of the ZnT family decrease the cytosolic concentration(Reference Guerinot19–Reference Palmiter and Huang22). Zip4 and ZnT1 are involved in Zn absorption in the small intestine(Reference McMahon and Cousins23, Reference Wang, Zhou and Kuo24), whereas Zip5 is responsible for intestinal Zn secretion(Reference Dufner-Beattie, Kuo and Gitschier25). In addition, Zn is secreted from the pancreas into the gut by Zip5 and ZnT1(Reference Dufner-Beattie, Kuo and Gitschier25, Reference Liuzzi, Bobo and Lichten26). Zn transporter activities are modulated in response to Zn depletion through alteration of gene expression, transporter translocation, or both(Reference Palmiter and Huang22).
We hypothesised that growth promotion induced by excess Zn intake (3000 mg/kg) is not limited to weaning piglets but is instead observed in other animals, including rats. In addition, it was hypothesised that each tissue grows proportionally in rats fed diets supplemented with excess Zn. In order to examine these points, we examined body and tissue growth, accumulation of Zn in tissues and expression of Zn transporters in growing rats fed diets supplemented with excess Zn. We specifically examined (1) whether growth promotion induced by excess Zn is observed in growing rats, (2) whether the concept proposed by the NRC regarding the prevention from Zn toxicity is applicable and (3) whether Zn transporter gene expression is altered in response to excess Zn ingestion. Our results indicate that excess Zn ingestion did not enhance the growth performance in growing rats, but in fact decreased pancreatic weight. Unexpectedly, the gene transcript levels of both the intestinal Zn transporters involved in Zn absorption and those involved in secretion were decreased in rats fed diets with higher Zn contents.
Materials and methods
Animals and diets
The experiments were approved by the Kyoto University Animal Experiment Committee (20-19). A total of twenty-eight male specific pathogen-free Sprague–Dawley rats aged 4 weeks were housed individually in stainless-steel cages under constant conditions (24°C, 50 % humidity) with a fixed light–dark cycle (lights on from 05.00 to 19.00 hours). Because excess Zn-induced growth promotion in weaning pigs is not necessarily due to the antimicrobial effects as described earlier, we used specific pathogen-free rats. After a 5 d acclimatisation period of feeding the basal diet (24 mg Zn/kg) shown in Table 1, rats were randomly assigned to receive diets with differing Zn concentrations. The requirement of Zn in growing rats is 12 mg/kg(27); the basal diet contained twice as much Zn(27). Zn content in the diet recommended for growing rats by AIN(Reference Reeves28) is 30 mg/kg. All groups were allowed free access to food and distilled water for the 10 d study period. The diets were prepared by addition of ZnO to a Zn-deficient diet at the expense of glucose to provide a 24, 1016, 2008 or 3000 mg Zn/kg diet, and the actual measured content was 23·8, 1050, 2090 or 3200 mg Zn/kg diet, respectively. Since this study tested responses to dietary Zn status, i.e. excess intake as well as deficiency, egg-white was used as a protein source. d-Biotin was added to the basal diet due to the high avidin content of egg-white in order to prevent biotin deficiency. Body weight (BW) and feed consumption were measured every day.
Table 1 Ingredients of the basal diet
* The vitamin mixture (g/kg) contains: nicotinic acid, 3·000; Ca pantothenate, 1·600; pyridoxine-HCl, 0·0700; thiamine-HCl, 0·600; riboflavin, 0·600; folic acid, 0·20; d-biotin, 0·020; vitamin B12 (cyanocobalamin) (0·1 % in mannitol), 2·500; vitamin E (all-rac-α-tocopheryl acetate) (500 mg/g), 15·00; vitamin A (all-trans-retinyl palmitate)(150 mg/g), 0·800; vitamin D3 (cholecalciferol) (10 mg/g), 0·250; vitamin K (phylloquinone), 0·075; powdered sucrose, 974·655.
† The mineral mixture (g/kg) contains: CaCO3, 357·00; KH2PO4, 196·00; K3(C6H5O7).H2O, 70·78; NaCl, 74·00; K2SO4, 46·60; MgO, 24·00; Fe(C6H5O7), 6·06; MnCO3, 0·63; CuCO3, 0·30; KIO3, 0·01; Na2O3Se, 0·01 025; (NH4)6Mo7O24.4H2O, 0·00 795; NaSiO2.9H2O, 1·45; CrK(SO4)2.12H2O, 0·275; LiCl, 0174; H3BO3, 0·0815; NaF, 0·0635; NiCO3, 0·0318; NH4VO3, 0·0066; powdered sucrose 222·676.
At the end of the 10 d experimental period, rats were killed by bleeding from the abdominal aorta under isoflurane anaesthesia. Tissues (the liver, kidney, pancreas, spleen, small intestine, testis, gastrocnemius muscle, femur and perirenal fat pad) were collected and weighed. Blood collected with a heparinised syringe was centrifuged at 2500 g for 30 min at 4°C to obtain the plasma. The intestine was flushed with saline and scraped with slide glass to obtain the intestinal epithelium. Other tissues were rinsed in saline, immediately frozen in liquid N2 and stored at − 80°C until analysis.
Determination of mineral contents
After wet-ash digestion of diets, plasma, and tissues with trace-element grade nitric acid and H2O2, Zn concentrations in diets, plasma and tissues were measured by atomic absorption spectrometry (AA-6600F; Shimadzu, Kyoto, Japan). The analytical accuracy of the Zn determination was confirmed by analysis of a certified reference material from bovine liver (standard reference material 1577b, National Institute of Standards and Technology, Gaithersburg, MD, USA).
RNA extraction and quantitative RT-PCR
Total RNA was isolated from the small intestine epithelium, pancreas and liver using TRIzol reagent (Invitrogen, Carlsbad, CA, USA), according to the manufacturer's protocol. Recovered RNA was used as a template for RT using random primers (ABI high-capacity complementary DNA reverse transcription kit; Applied Biosystems, Foster City, CA, USA). The quantitative RT-PCR was carried out using an SYBR premix Ex Taq II kit (TaKaRa, Otsu, Japan) in a Roter-Gene 6000 instrument (Corbett Research, Mortlake, Australia). PCR was performed as follows: an initial denaturation step of 10 s at 95°C, followed by forty cycles of 5 s at 95°C and 20 s at 60°C. The dissociation (melting) curve of quantitative RT-PCR products was subsequently examined by changing the ramp temperature from 60°C to 94°C. Each sample showed a single peak, suggesting that the expected PCR products were obtained. The PCR primers used to detect Zip4, Zip5, ZnT1, metallothionein-1a (Mt-1a), Mt-2a, insulin-like growth factor-1 (Igf-1) and hypoxanthine phosphoribosyltransferase 1 (Hprt1) were as follows: 5′-AAC CCA CCA GGG AGG AGA-3′ and 5′-TTC TGG AAA CCC CTG CTT C′-3′ for Zip4; 5′-CCT CGG GCC TAG ACC TCT T-3′ and 5′-AGC TGG GAA CCA TTC AGA CA-3′ for Zip5; 5′-AAC ACC AGC AAT TCC AAC G-3′ and 5′-CCA CTG GAT CAT CAC TTC TCA A-3′ for ZnT1; 5′-CAC CAG ATC TCG GAA TGG AC-3′ and 5′-GCA GCA GCT CTT CTT GCA G-3′ for Mt-1a; 5′-ACC TCC TGC AAG AAA AGC TG-3′ and 5′-ACT TGT CCG AAG CCT CTT TG-3′ for Mt-2a; 5′-GGA CGC TCT TCA ATT CGT GT-3′ and 5′-CTT CAG CGG AGC ACA GTA CA-3′ for Igf-1; and 5′-GAC CGG TTC TGT CAT GTC G-3′ and 5′-ACC TGG TTC ATC ATC ACT AAT CAC-3′ for Hprt1. Gene transcript levels in each sample were determined using the relative standard curve method. The level of gene transcripts was expressed as a ratio relative to Hprt1 mRNA, with the level in rats fed the basal diet set to 1.
Statistical analyses
Data are expressed as the least square means with their standard errors. All analyses were performed using SAS (SAS Institute, Cary, NC, USA)(29). The data on BW and feed intake were subjected to the MIXED procedure. Each rat was determined an experimental unit and measurements of the same rat on different days were considered repeated measures. The statistical model included the effects of diet, experimental day and the interaction between both. In addition, the effects of dietary Zn on tissue weight were analysed with the general linear model procedure. Furthermore, when tissue Zn concentrations or weights were plotted against daily Zn intake, a breaking point of daily Zn intake indicating a plateau was explored using the NLIN procedure. A model with 1 breaking point and no limit of its slope value before or after the point was applied. When the model was significant, and when the slope was not significantly different from 0, the breaking point was further determined by application of the model with 1 breaking point with 0 as the slope after the point. Differences were considered significant at P < 0·05.
Results
Body and tissue growth
Time-course changes in BW showed an insignificant effect of diet, but the effect of the interaction between diet and experimental day was significant, suggesting that the effect of excess Zn on BW depended on the length of the treatment period (Fig. 1). This indicates that, unlike in piglets, a diet supplemented with 3000 mg Zn/kg does not have a beneficial effect on BW gain in rats. As for daily feed intake, both the diet effect and the interaction between diet and experimental day were not statistically significant (Fig. 2(A)). In addition, feed efficiency, i.e. weight gain per feed intake, was not significantly affected by the diet, the experimental day or the interaction between both (Fig. 2(B)).
Fig. 1 Time-course changes in body weight (BW) in growing rats. Rats were fed diets containing various concentrations of Zn for 10 d. BW was plotted against experimental days. Values are means with their standard errors represented by vertical bars (n 7). 
, 24 mg Zn/kg; 
, 1016 mg Zn/kg; 
, 2008 mg Zn/kg; 
, 3000 mg Zn/kg.
Fig. 2 Time-course changes in daily feed intake and feed efficiency in growing rats. Rats were fed diets containing various concentrations of Zn for 10 d. (A) Daily feed intake and (B) feed efficiency were plotted against experimental days. Values are means with their standard errors represented by vertical bars (n 7). , 24 mg Zn/kg; , 1016 mg Zn/kg; , 2008 mg Zn/kg; , 3000 mg Zn/kg.
Typical symptoms of Zn toxicity are vomiting and gastrointestinal dysfunction(Reference Fosmire30), and the pancreas is the organ most sensitive to Zn toxicity(Reference Sutomo, Woutersen and Van den Hamer6). Two rats fed a diet supplemented with 3000 mg Zn/kg excreted soft stools for the last 3 d of the experimental period, although the other rats did not exhibit any symptoms throughout the study. The weight of the pancreas relative to BW was lower in rats fed a diet supplemented with 2008 mg Zn/kg or 3000 mg Zn/kg compared to those fed a diet supplemented with 24 mg Zn/kg or 1016 mg Zn/kg, whereas no significant effects of the diet on the weight of the liver, kidney, spleen, testis, gastrocnemius muscle, femur and perirenal fat were detected (Table 2). Plotting the pancreas weight against daily Zn intake revealed a decrease in rats that ingested Zn at levels above 15·2 mg/d (Fig. 3).
Table 2 Relative tissue weight of rats fed the diets supplemented with excess zinc
(Mean values with their standard errors, n 7)
a,b Mean values with unlike letters were significantly different (P < 0·05).
* Average weight of tissues from both left and right parts.
Fig. 3 Effects of excess Zn intake on the weight of the pancreas in growing rats. Rats were fed diets with various concentrations of Zn for 10 d. The weight of the pancreas relative to body weight (BW) was plotted against average daily intake of Zn. The breaking point of daily Zn intake on the pancreas weight was calculated, and shown in the figure by an arrow. ○, 24 mg Zn/kg; 
, 1016 mg Zn/kg; 
, 2008 mg Zn/kg; 
, 3000 mg Zn/kg.
Plasma and tissue zinc concentration
Plasma and tissue Zn concentrations were higher in rats fed diets supplemented with more than 1016 mg Zn/kg than in those fed the basal diet (data not shown). To examine the relationship between Zn intake and the Zn concentrations in detail, plasma and tissue Zn concentrations were plotted against daily Zn intake (Fig. 4). Zn concentrations in the plasma, liver, kidney and femur increased linearly with increasing intake of Zn up to 31·3, 37·1, 28·3 and 35·1 mg/d, respectively, and eventually reached a plateau (Fig. 4(A)–(D)). These results suggest that Zn efficiently accumulates in the plasma, liver, kidney and femur in response to increased Zn intake and that the capacity to retain Zn is limited. In view of the higher proportion of skeletal muscle weight relative to total BW, approximately 60 % of body Zn is stored in the muscle tissue(Reference Jackson and Mills31); however, the Zn concentration in the muscle was not significantly altered by increasing Zn intake (Fig. 4(E)).
Fig. 4 Relationship between Zn intake and plasma and tissue concentrations of Zn in growing rats. Rats were fed diets with various concentrations of Zn for 10 d. Zn concentrations in the (A) plasma, (B) liver, (C) kidney, (D) femur, (E) gastrocnemius muscle and (F) pancreas were plotted against average daily intake of Zn. The breaking point of daily Zn intake on plasma and tissue concentrations of Zn was calculated, and indicated in the figure by an arrow. ○, 24 mg Zn/kg; , 1016 mg Zn/kg; , 2008 mg Zn/kg; , 3000 mg Zn/kg. Note that there was no break point on Zn concentration in the gastrocnemius muscle.
The relationship between the Zn concentration in the pancreas and daily Zn intake also indicated that the breaking point was 8·4 mg/d; the pancreatic Zn concentration increased up to this point (Fig. 4(F)). In contrast to the plasma, liver, kidney and femur, the Zn concentration in the pancreas linearly decreased in rats that ingested Zn above the breaking point.
Expression of zinc transporters, metallothionein and insulin-like growth factor-1
The gene transcript levels of the transporters involved in intestinal absorption and secretion of Zn were examined by quantitative RT-PCR (Fig. 5). Expression of Zip4 in the intestinal epithelium, which is responsible for the uptake of Zn across the mucosal membrane(Reference Wang, Zhou and Kuo24), was significantly lower in rats fed a diet supplemented with 1016 mg Zn/kg than in those fed the basal diet, with no differences among dietary groups with higher Zn contents (Fig. 5(A)). The expression of ZnT1 in the intestinal epithelium, which promotes Zn transport from the cytosol of epithelial cells to the portal vein(Reference McMahon and Cousins23), was also down-regulated in rats fed diets supplemented with 1016–3000 mg Zn/kg (Fig. 5(B)). Zip5 is located at the basolateral membrane of the intestinal epithelium and promotes Zn transport from the portal vein to epithelial cells(Reference Dufner-Beattie, Kuo and Gitschier25). The expression of Zip5 was down-regulated in rats fed diets supplemented with 1016–3000 mg Zn/kg; the expression in rats fed a diet supplemented with 2008 mg Zn/kg was higher than that in rats fed a diet containing 1016 mg Zn/kg (Fig. 5(C)). No significant differences were detected in the expression of ZnT1 and Zip5, transporters responsible for Zn secretion into the gut(Reference Dufner-Beattie, Kuo and Gitschier25, Reference Liuzzi, Bobo and Lichten26), in the pancreas (Fig. 5(D) and (E)).
Fig. 5 Gene expression of Zn transporters in the small intestinal epithelium and the pancreas of rats. Rats were fed diets supplemented with various concentrations of Zn for 10 d. Gene expression of (A) Zip4, (B) ZnT1 and (C) Zip5 in the small intestine, and (D) ZnT1 and (E) Zip5 in the pancreas was examined by quantitative RT-PCR. The transcription levels were expressed as ratios to hypoxanthine phosphoribosyltransferase 1 with the level in rats fed the basal diet set to 1. Values are means with their standard errors represented by vertical bars (n 7). a,b,c Mean values with unlike letters were significantly different (P < 0·05).
Mt is involved in Zn homeostasis, and Mt expression is induced by several metals including Zn(Reference Palmiter32, Reference Mocchegiani, Malavolta and Costarelli33). Expression of Mt-1a has been shown to change in parallel with that of ZnT1 in response to Zn exposure in cultured hepatoma cells and fibroblasts(Reference Langmade, Ravindra and Daniels34); we therefore evaluated Mt expression in the present study. The expression levels of Mt-1a and Mt-2a in the intestinal epithelium were higher in rats fed diets supplemented with 2008 or 3000 mg Zn/kg than in those fed the basal diet (Fig. 6(A) and (B)); Mt-1a expression was higher in rats fed a diet supplemented with 2008 mg Zn/kg than in those fed a diet containing 3000 mg Zn/kg. The expression of Mt-1a and Mt-2a in the liver was higher in rats fed diets supplemented with 1016–3000 mg Zn/kg than in those fed the basal diet; the expression increased in a dose-dependent manner (Fig. 6(C) and (D)). The regulatory expression of hepatic Mt-1a and Mt-2a in response to excess Zn intake is similar to the changes in Zn accumulation in the liver (Fig. 4(B)); this reflects the fact that Zn accumulated in the liver is incorporated into Mt-1a and Mt-2a, which buffer excess Zn to provide protection from Zn toxicity(Reference Mocchegiani, Malavolta and Costarelli33). No significant changes in Mt expression were detected in the pancreas (Fig. 6(E) and (F)).
Fig. 6 Gene expression of metallothionein (Mt) in the small intestinal epithelium, liver and pancreas of rats. Rats were fed diets supplemented with various concentrations of Zn for 10 d. Gene expression of (A, C, E) Mt-1a and (B, D, F) Mt-2a in the (A, B) small intestine, (C, D) liver and (E, F) pancreas was examined by quantitative RT-PCR. The transcription levels were expressed as ratios to hypoxanthine phosphoribosyltransferase 1 with the level in rats fed the basal diet set to 1. Values are means with their standard errors represented by vertical bars (n 7). a,b,c Mean values with unlike letters were significantly different (P < 0·05).
A previous study suggested that the increased expression of Igf-1 and its receptor are responsible for excess Zn-induced growth promotion in piglets(Reference Li, Yin and Li13). We evaluated expression of Igf-1 in the intestinal epithelium and in the liver, one of the major Igf-1-producing organs(Reference Ohlsson, Mohan and Sjögren35). Expression of Igf-1 was not significantly different among groups, irrespective of the tissues analysed (Fig. 7).
Fig. 7 Gene expression of insulin-like growth factor-1 (Igf-1) in the small intestinal epithelium and liver of rats. Rats were fed diets supplemented with various concentrations of Zn for 10 d. Gene expression of Igf-1 in the (A) small intestine and (B) liver was examined by quantitative RT-PCR. The transcription levels were expressed as ratios to hypoxanthine phosphoribosyltransferase 1 with the level in rats fed the basal diet set to 1. Values are means with their standard errors represented by vertical bars (n 7).
Discussion
The present study examined the short-term effects of extremely high Zn intake on body and tissue weight; tissue Zn concentration; and the expression of Zn transporters, Mt, and Igf-1 in growing rats. Our results revealed that, in contrast to studies in piglets, excess intake of Zn did not have any beneficial effects on growth, but rather induced a decrease in weight of the pancreas. The NRC(7) proposed that excess Zn predominantly accumulates in organs such as the liver, kidney and bone in order to protect the pancreas, the organ most sensitive to excess Zn. The present results basically support this model: Zn concentrations in the liver, kidney and femur proportionally increased with increasing Zn intake up to 28–37 mg/d. However, the weight of the pancreas was lower in rats that ingested more than 15 mg/d Zn. This indicates that the amount of Zn intake required to induce atrophy of the pancreas is smaller than that required to reach a plateau in Zn accumulation in other tissues. Thus, the present results obtained in rats ingesting extremely high concentrations of Zn for a short period suggest the imperfect buffering capacity of the liver, kidney and bone against excess Zn ingestion. Furthermore, our results suggest that Zn accumulation in the skeletal muscle, the tissue that stores the majority of Zn in the body, is not affected by dietary Zn intake.
Zn depletion-induced Zip4 expression in the small intestine is well-characterised(Reference Dufner-Beattie, Wang and Kuo36–Reference Jou, Hall and Philipps38). By contrast, less information is available regarding the expression of Zn transporters in animals fed excess Zn. Expression of ZnT1 and ZnT2 in the small intestine was significantly higher in rats fed a diet containing 180 mg Zn/kg for 2 weeks than in those fed a diet containing 30 mg Zn/kg(Reference Liuzzi, Blanchard and Cousins39). The present study indicates that excess Zn ingestion clearly down-regulates the mRNA expression of the Zn transporters involved in intestinal Zn absorption, Zip4 and ZnT1. The decrease in Zip4 expression was particularly evident; the gene transcript level of Zip4 in rats fed a diet supplemented with 1016 mg Zn/kg was only around 5 % of that in rats fed the basal diet, whereas the ZnT1 mRNA level of rats fed a diet containing 1016 mg Zn/kg was around 35 % of that in rats fed the basal diet. These results suggest that in addition to a system detecting Zn depletion, intestinal cells also have a system to sense excess Zn, and they partly regulate Zn absorption through transcriptional inhibition of Zn transporters.
ZnT1, Mt-1a and Mt-2a are transcriptionally regulated by metal-responsive transcription factor-1, a Zn-sensing transcription factor(Reference Laity and Andrews40), and the expression of both genes increases in response to Zn exposure in cultured cells(Reference Langmade, Ravindra and Daniels34, Reference Yang, Kroft and Chitambar41, Reference Jackson, Valentine and McKay42); however, excess Zn ingestion (>2008 mg Zn/kg) caused the down-regulation of ZnT1 mRNA expression in the intestinal epithelium but up-regulation of Mt-1a and Mt-2a mRNA expression. An unidentified additional regulatory mechanism of ZnT1 expression is probably involved in rats fed a diet with extremely high Zn contents.
The capacity for Zn accumulation in the liver, kidney and bone in response to excess Zn ingestion was limited. Rats that ingested 28–37 mg Zn/d could not accumulate additional Zn in these tissues; this Zn level corresponded to ingestion of a diet supplemented with 2008 mg Zn/kg. These results suggest stimulation of Zn secretion by feeding diets containing more than 2008 mg Zn/kg, or inhibition of Zn absorption, or both. Although Zip5 plays a role in Zn secretion across the intestinal mucosa(Reference Dufner-Beattie, Kuo and Gitschier25), the gene transcript level of Zip5 in the intestinal epithelium was not elevated but instead decreased in rats fed a diet supplemented with 1016 mg Zn/kg compared to those fed the basal diet. Thus, it is unlikely that the Zip5-mediated Zn secretion is increased in rats fed diets with high Zn content. The pancreas is the major organ for endogenous Zn secretion into the gut(Reference McClain43). Expression levels of pancreatic Zip5 and ZnT1, which are expressed predominantly in the acinar cells of the pancreas(Reference Dufner-Beattie, Kuo and Gitschier25, Reference Liuzzi, Bobo and Lichten26), were not increased in response to the ingestion of diets supplemented with excess Zn. These results suggest that the increased Zn secretion from the pancreas is not responsible for the limited Zn accumulation in the liver, kidney and bone.
Intestinal Mt levels are inversely related to the rate of Zn absorption(Reference Hoadley, Leinart and Cousins44), and it has been suggested that Mt inhibits Zn absorption(Reference Davis, McMahon and Cousins45). Thus, the present finding that expression of Mt-1a and Mt-2a was up-regulated in the intestinal epithelium of rats fed diets supplemented with 2008 or 3000 mg Zn/kg may reflect an inhibition of intestinal Zn absorption in these rats. However, the expression of intestinal Mt-1a and Mt-2a was not significantly higher in rats fed the diet supplemented with 3000 mg Zn/kg than in those fed the diet containing 2008 mg Zn/kg. Thus, the limit of tissue Zn accumulation in rats fed the diet supplemented with 2008 mg Zn/kg could not be explained by the Mt-induced inhibition of Zn absorption. Although gene transcript levels of intestinal Zip4 were not further decreased in rats fed diets containing more than 1016 mg Zn/kg in this study, processing and translocation of Zip4 are modified in response to changes in Zn status(Reference Kim, Wang and Dufner-Beattie46, Reference Kambe and Andrews47). Thus, post-translational modifications may be responsible for the defence against Zn ingestion in diets containing more than 2008 mg Zn/kg.
In piglets, short-term feeding of a diet supplemented with 3000 mg Zn/kg effectively enhanced BW gain(Reference Hill, Mahan and Carter11–Reference Li, Yin and Li13), and this feeding regimen is applied in practice on pig farms. Li et al. (Reference Li, Yin and Li13) suggested that the excess Zn ingestion stimulates Igf-1-mediated signalling, which enhances the villous height of the small intestinal mucosa resulting in growth promotion of piglets. Considering that Igf-1 expression in the small intestine and in the liver was not significantly increased in response to excess Zn ingestion, the inability to stimulate the Igf-1 axis may be one of the reasons why body growth was not accelerated in rats fed a diet supplemented with 3000 mg Zn/kg.
The present study clarified that growing rats have defence mechanisms against excess Zn ingestion; in addition to the effective accumulation of excess Zn in the liver, kidney and bone, the down-regulated mRNA expression of Zn transporters involved in intestinal absorption contributes to the protection of the pancreas against excess Zn-mediated adverse effects. Furthermore, up-regulation of Mt expression in the small intestine results in the inhibition of Zn absorption. These multiple defences may contribute to the relative tolerance to excess Zn ingestion.
Acknowledgements
The present study received no specific grant from any funding agency in the public, commercial or not-for-profit sectors. T. F., T. M. and M. F. designed the experiments; T. F. performed the experiments; T. F., T. M. and M. F. analysed the data and wrote the paper. All authors discussed the results and approved the manuscript in its final version. The authors declare no conflict of interest.
