Cadmium transport by human Nramp 2 expressed in Xenopus laevis oocytes

Available online at www.sciencedirect.com R

Toxicology and Applied Pharmacology 187 (2003) 162–167

www.elsevier.com/locate/taap

Cadmium transport by human Nramp 2 expressed in Xenopus laevis oocytes Masato Okubo,a Kyohei Yamada,a Makoto Hosoyamada,b,* Toshiaki Shibasaki,a and Hitoshi Endoub a

b

Department of Therapeutics, Kyoritsu College of Pharmacy, Shibakoen 1-5-30, Minato-ku, Tokyo, 105-8512, Japan Department of Pharmacology and Toxicology, Kyorin University School of Medicine, Shinkawa 6-20-2, Mitaka, Tokyo, 181-8611, Japan Received 22 August 2002; accepted 27 November 2002

Abstract Using the Xenopus oocyte expression system, human Nramp2, a human intestinal iron transporter, was shown to work as a cadmium transporter. An 1824-bp human Nramp2 cDNA was constructed by PCR cloning from reverse transcription products of human kidney mRNA. When the pH of the extracellular solution was 6.0, human Nramp2 transported 109Cd2⫹. Substitution of external Cl⫺ with NO3⫺ had no effect on human Nramp2-dependent cadmium uptake. The concentration-dependent Cd2⫹ transport of human Nramp2 indicated Michaelis–Menten type transport with an average Km value of 1.04 ⫾ 0.13 ␮M and an average Vmax of 14.7 ⫾ 1.9 pmol/oocyte/h (n ⫽ 3). Cd2⫹ transport via human Nramp2 was inhibited significantly by Cd2⫹, Fe2⫹, Pb2⫹, Mn2⫹, Cu2⫹, and Ni2⫹, while it was not inhibited by Hg2⫹ and Zn2⫹. Transport of 0.1 ␮M Cd2⫹ by human Nramp2 was inhibited by metallothionein (IC50 ⫽ 0.14 ␮M). Therefore, human Nramp2 is suggested to function as a pH-dependent cadmium absorption transporter on the luminal membrane of human intestinal cells. © 2003 Elsevier Science (USA). All rights reserved. Keywords: Metal transport; Cadmium; Nramp2; Iron; Zinc; Mercury; Metallothionein

Introduction Cadmium (Cd) is a highly toxic metal and has been recognized as an environmental contaminant that may enter the food chain. Thus, the mechanisms of intestinal cadmium absorption need to be elucidated for understanding cadmium toxicity to the human body. Since it is assumed that the mammalian organism has not developed specialized intestinal absorption mechanisms for nonessential metals, it is generally considered that uptake of cadmium occurs via pathways also used by essential trace metals. Intestinal uptake of cadmium is inhibited by zinc (Foulkes, 1991) and iron (Sullivan and Ruemmler, 1987). Natural resistanceassociated macrophage protein 2 (Nramp2) is thought to be the main transporter that absorbs iron at the brush-border membrane of mammalian intestine (Andrews, 2000). Rat Nramp2, also designated as DCT1 or DMT1, was identified * Corresponding author. Fax: ⫹81-422-79-1321. E-mail address: [email protected] (M. Hosoyamada).

as a proton-coupled metal ion transporter, which demonstrated Fe2⫹, Zn2⫹, and Cd2⫹ transport electrophysiologically (Gunshin et al., 1997). Iron transport by human Nramp2 expressed in Chinese hamster ovary (CHO) cells has been measured using the fluorescent, metal-sensitive dye calcein and was inhibited by Cd2⫹ (Picard et al., 2000). Human Nramp2 was reported to be localized to the brush border of human duodenum by immunohistochemistry (Griffiths et al., 2000). Up-regulated Nramp2 expression in iron-deficiency status (Han et al., 1999) is coincident with increased cadmium absorption in humans with iron deficiency (Flanagan et al., 1978). Therefore, human Nramp2 is a candidate for the cadmium transporter at the brush border of human duodenum. Elisma and Jumarie (2001) reported that the specific uptake of 109Cd by human enterocyte-like Caco-2 cells increased fourfold as extracellular pH was lowered from 7.5 to 5.5. Furthermore, this 109Cd transport was characterized as a saturable system (Km ⫽ 1.1 ⫾ 0.1 ␮M, Vmax ⫽ 87 ⫾ 3 pmol/3 min/mg protein). An excess of Fe2⫹ failed to

0041-008X/03/$ – see front matter © 2003 Elsevier Science (USA). All rights reserved. doi:10.1016/S0041-008X(02)00078-9

M. Okubo et al. / Toxicology and Applied Pharmacology 187 (2003) 162–167

affect 109Cd uptake when the extracellular pH was 7.4, whereas a strong inhibition was observed when the extracellular pH was 5.5. In contrast, the maximal inhibitory effect of Zn2⫹ was observed when extracellular pH was 7.4. Therefore, the authors concluded that Fe2⫹ might compete with Cd2⫹ for Nramp2, whereas Zn2⫹ competes for a different system of cadmium transport in the Caco-2 cell. The purpose of the present study is to confirm and characterize 109Cd transport by human Nramp2 molecule in acidified extracellular conditions using the Xenopus oocyte expression system. The results suggest that human Nramp2 is one of the transporters that absorb cadmium at the brushborder membrane of the human intestine.

Methods Construction of human Nramp2 cDNA. Oligonucleotide primers (5⬘-GGC GGC GTG TCA GGT GGT TGC GGA GCT GGT AAG AAT CAT AT-3⬘ and 5⬘-GCA GGT AGC CAT CAG AGC CAG TGT GTT TCT ATG GTT TAC TGT GTG-3⬘) based on the human Nramp2 sequence (Genbank Accession no. NM000617) were used for RT–PCR amplification with the Advantage-HF2 PCR kit (BD Biosciences Clontech, Palo Alto, CA) from human kidney poly(A)⫹ RNA (BD Biosciences Clontech), which had been reverse transcribed ahead by the Advantage-2 reverse transcription kit (BD Biosciences Clontech). The 1824-bp PCR product of human Nramp2 cDNA was subcloned using a TA cloning kit (Invitrogen Co., Carlsbad, CA). The sequence of the isolated clone was confirmed by the BigDye Terminator cycle sequencing kit (Applied Biosystems, Foster City, CA) with synthetic oligonucleotide primers using the automated ABI 310 DNA sequencer. The Nramp2 cDNA was excised using EcoRV and KpnI and was subcloned into the EcoRV–KpnI site of pcDNA3.1(⫹) (Invitrogen). Analysis of Cd transport by human Nramp2 expressed in Xenopus oocytes. Human Nramp2 RNA was synthesized using mMESSAGE mMACHINE kit (Ambion, Austin, TX) from the linearized Nramp2 cDNA by HindIII. Poly(A)⫹ tail was added using Poly(A) Tailing kit (Ambion). Xenopus oocytes were excised from female Xenopus laevis (Saitama Experimental Animal, Saitama, Japan) anesthetized with 0.1% tricaine/0.3% KHCO3 and were defolliculated with 2 mg/ml collagenase type IA (Sigma-Aldrich Co., St. Louis, MO) in OR2 containing (in mM) 82 NaCl, 2.5 KCl, 1.0 MgCl2, and 5.0 Hepes, pH 7.5 for 2 h at room temperature. Defolliculated oocytes were injected with 30 –50 ng human Nramp2 RNA and incubated in Barth’s solution containing (in mM) 88 NaCl, 1.0 KCl, 0.74 CaCl2/NO3, 0.82 MgSO4, 2.4 NaHCO3, and 10 Hepes, pH7.4, for 3 days at 18°C until uptake studies were performed. Uptake studies were performed at room temperature in ND96-pH7.4 containing (in mM) 96 NaCl, 2.0 KCl, 1.8

163

CaCl2, 1.0 MgCl2, and 5.0 Hepes, pH7.4 or ND96-pH6.0 (Hepes was substituted with MES, pH 6.0) containing 2.2– 3.3 nM 109CdCl2 (1.85–37 MBq/␮g Cd; Amersham Bioscience Co., Piscataway, NJ) and various concentrations of cold CdCl2. NaCl in ND96 solution was substituted with NaNO3 in some experiments as NaNO3-ND96 solution. For the inhibition studies, 10 ␮M of each of the divalent metal ions (Cd2⫹, Fe2⫹, Pb2⫹, Mn2⫹, Cu2⫹, Ni2⫹, Hg2⫹, and Zn2⫹) was added to the uptake solution, 3.3 nM 109CdCl2/ ND96, pH 6.0. Metallothionein (Sigma-Aldrich Co.) was mixed with 0.1 ␮M CdCl2/ND96 pH 6.0 containing 3.3 nM 109 CdCl2 for 1 h before the uptake studies for acceleration of binding of metallothionein and 109CdCl2. After uptake mixtures were stopped and washed five times with ice-cold 109 Cd-free ND96 pH7.4; each oocyte solubilized with 200 ␮l of 10% SDS was mixed with 2.5 ml of Aquasol-2 (Packard, Meriden, CT) for radioactivity determination using a scintillation counter (LC-3010, ALOKA, Japan). Statistical analysis. The experiments were performed using three batches of oocytes from different animals, and results from the representative experiments are expressed as means ⫾ SE. Statistical analyses were performed using the unpaired t test. A value of p ⬍ 0.05 was the cutoff for significance.

Results pH dependency and influence of NO3 substitution of human Nramp2-dependent cadmium transport Figure 1 indicates the influence of extracellular pH and NO3⫺ anion on cadmium transport using human Nramp2 RNA-injected and noninjected X. laevis oocytes. When the pH of the extracellular solution was 7.4, the 109Cd transport by human Nramp2 RNA-injected oocytes did not differ from the intrinsic 109Cd transport by noninjected oocytes. Intrinsic 109Cd transport by noninjected oocytes was attenuated at an extracellular pH of 6.0, whereas 109Cd transport by human Nramp2 RNA-injected oocytes was significantly enhanced in the same extracellular solution. Therefore, human Nramp2-dependent cadmium transport was proton driven, similar to ferrous iron transport by rat Nramp2/ DCT1/DMT1, and the human Nramp2 molecule is concluded to be a cadmium transporter. When the pH of the extracellular solution was 6.0, 109Cd transport by human Nramp2 RNA-injected oocytes in the NaCl-ND96 solution was not different from that in the NaNO3-ND96 solution. However, intrinsic 109Cd transport by noninjected oocytes in the NaNO3-ND96 was higher than that in the NaCl-ND96 solution. Thus, the human Nramp2-dependent cadmium transport was not influenced by the substitution of extracellular Cl⫺ anion with NO3⫺ anion. In contrast, the X. laevis oocyte contained the intrin-

164

M. Okubo et al. / Toxicology and Applied Pharmacology 187 (2003) 162–167

Fig. 1. Transport of 3.3 nM 109Cd in noninjected (open columns) or human Nramp2 RNA-injected (filled columns) Xenopus oocyte under ND96-NaCl/pH 7.4, or pH 6.0, and ND96-NaNO3/pH 7.4 or pH 6.0 as described under Methods. Points shown are means ⫾ SE (n ⫽ 8). **p ⬎ 0.01 vs noninjected oocytes in the same solution.

sic cadmium transport pathway that was influenced by the substitution of extracellular Cl⫺ anion with NO3⫺ anion. Time profile and concentration dependence of human Nramp2-dependent cadmium transport 109

Cd uptake by human Nramp2 RNA-injected oocytes increased linearly in a time-dependent manner for 45 min

and was significantly different from intrinsic 109Cd uptake by noninjected oocytes after 15 min (Fig. 2). Therefore, we observed 109Cd uptake for either 15 or 30 min in the following experiments. Figure 3 indicates a typical plot of the concentration dependence of human Nramp2-dependent cadmium uptake derived by the subtraction of 109Cd transport by noninjected oocytes from the 109Cd transport by human Nramp2 RNA-

Fig. 2. Time course of 0.3 ␮M 109Cd uptake in noninjected (open circles) or human Nramp2 RNA-injected (filled circles) Xenopus oocyte under ND96-NaCl pH 6.0. Points shown are means ⫾ SE (n ⫽ 8). **p ⬎ 0.01 vs noninjected oocytes.

M. Okubo et al. / Toxicology and Applied Pharmacology 187 (2003) 162–167

165

Fig. 3. Typical figure of concentration dependence of 109Cd transport by human Nramp2 expressed in Xenopus oocyte under ND96-NaCl pH 6.0. (Inset) Eadie–Hofstee plot (r2 ⫽ 0.9832). Points shown are means ⫾ SE (n ⫽ 8). From three independent experiments, Km value was calculated as 1.04 ⫾ 0.13 ␮M and Vmax was calculated as 14.7 ⫾ 1.9 pmol/oocyte/h.

injected oocytes. From this figure, it can be seen that human Nramp2-dependent cadmium uptake was Michaelis–Menten type transport, and its Eadie–Hofstee plot, shown in the inset in the same figure, is related in a linear fashion. The Km value, averaged from three independent experiments, was 1.04 ⫾ 0.13 ␮M and the average Vmax was 14.7 ⫾ 1.9 pmol/oocyte/h.

metallothionein for 1 h. The inhibition by metallothionein was significant at concentrations above 0.3 ␮M and the IC50 was 0.14 ␮M (Fig. 5).

Inhibition of human Nramp2-dependent cadmium transport by divalent cations and metallothionein

Cadmium transport by human Nramp2 was not observed when extracellular pH was 7.4. Therefore, cadmium uptake by Caco-2 cells (Elisma and Jumarie, 2001) at pH 7.4 did not respond to human Nramp2-mediated Cd2⫹ uptake. Orally administrated cadmium chloride was found to be deposited in the duodenum of rats (Phillpotts, 1979) and mice (Sorensen et al., 1993). Since intraluminal pH of human duodenum is reported to be between 5.0 and 6.0 at the duodenojejunal junction (Ovesen et al., 1986), human Nramp2 expressed on the brush-border membrane of human duodenum seems to transport cadmium in vivo. Cadmium uptake by Caco-2 cells in acidified extracellular NaNO3 solution was reported to be higher than that in acidified extracellular NaCl solution. Although cadmium uptake by noninjected oocytes was enhanced in acidified extracellular NaNO3 solution, cadmium uptake by human Nramp2 was not affected by substitution of NaCl with NaNO3. Since NO3⫺ anion was reported to induce intracellular acidification (Chow et al., 1997), extracellular NaNO3 solution might decrease the inward directed proton gradient across the cellular membrane, which is the driving force of human Nramp2-dependent transport. Like Xenopus oocytes,

Since rat DCT1/DMT1/Nramp2 is known to be a divalent cation transporter, the inhibition of human Nramp2-dependent cadmium transport was investigated by the addition of different divalent cations to the 109Cd uptake solution. As shown in Fig. 4, 10 ␮M CdCl2, FeCl2, Pb(CH3COO)2, MnCl2, CuCl2, and NiCl2 significantly inhibited the human Nramp2-dependent 109Cd transport to 7.9, 14.7, 22.6, 42.4, 58.1, and 62.1% of control, respectively. Therefore, these metal cations could inhibit human Nramp2-dependent cadmium transport at a high enough concentration. In contrast, the human Nramp2-dependent cadmium transport was not affected by 10 ␮M HgCl2 or 10 ␮M ZnCl2. Cadmium is usually bound to metallothionein in animal food products. Therefore, the inhibition of human Nramp2dependent cadmium transport by metallothionein was investigated by preincubation with metallothionein purified from rabbit liver. Human Nramp2-dependent 109Cd transport was attenuated in the uptake solution containing 0.1 ␮M cadmium preincubated with varying concentrations of

Discussion

166

M. Okubo et al. / Toxicology and Applied Pharmacology 187 (2003) 162–167

Fig. 4. Inhibition of human Nramp2-dependent 109Cd uptake by 10 ␮M divalent metal cations under ND96-NaCl pH 6.0. Values shown are means ⫹ SE of percentage uptake of human Nramp2-dependent 109Cd uptake without inhibitors as a control (18.7 ⫾ 1.7 fmol/oocyte/h, n ⫽ 8). **p ⬎ 0.01; *p ⬎ 0.05 vs no inhibitor.

Caco-2 cells may contain a cadmium transport pathway that is enhanced in acidified extracellular NaNO3 solution. The Km value of cadmium transport by human Nramp2 was almost the same as that observed in Caco-2 cells (K0.5 ⫽ 1.3 ⫾ 0.5 ␮M) (Elisma and Jumarie, 2001). This value is similar to values measured in Xenopus oocytes for DCT1/rat Nramp2 (Km ⫽ 1.1 ⫾ 0.1 ␮M, Gunshin et al., 1997) and yeast SMF1 (Km ⫽ 2.2 ⫾ 0.2 ␮M. Chen et al., 1999)

mediated iron transport. Since daily intake of cadmium was reported to be 15 ␮g (0.13 ␮mol. Jarup et al., 1998), intraluminal concentration of cadmium is calculated to be 0.1 ␮M when daily food intake is estimated to be 1.3 kg and no inhibitor of cadmium absorption exists in the food. Free cadmium at this concentration seems to be absorbed by Nramp2 in the human duodenum without saturation. Human Nramp2, which is expressed at the plasma mem-

Fig. 5. Inhibition of human Nramp2-dependent 109Cd uptake by various concentration of metallotionein under ND96-NaCl pH 6.0. Points shown are means ⫾ SE of percentage uptake of human Nramp2-dependent 0.1 ␮M 109Cd uptake without metallothionein as a control (1.45 ⫾ 0.53 pmol/oocyte/h, n ⫽ 8). **p ⬎ 0.01; *p ⬎ 0.05 vs the uptake in metallothionein-free ND96-NaCl pH 6.0.

M. Okubo et al. / Toxicology and Applied Pharmacology 187 (2003) 162–167

brane of CHO cells, transported iron and cobalt into the calcein-accesible cytoplasmic pool, and its iron transport was inhibited by Cd2⫹ (Picard et al. 2000). In our study, 10 ␮M Fe2⫹ inhibited 109Cd2⫹ uptake by human Nramp2 but 10 ␮M Zn2⫹ failed to inhibit it. These results corroborate those reported for Caco-2 cells; cadmium uptake by Caco-2 cells in acidified extracellular NaNO3 solution was inhibited by 100 ␮M Fe2⫹ but was not inhibited by 100 ␮M Zn2⫹ (Elisma and Jumarie, 2001). On the contrary, Zn2⫹ transport by rat DCT1/DMT1/Nramp2 was demonstrated electrophysiologically using 50 ␮M Zn2⫹ (Gunshin et al., 1997). Thus, Zn-sensitive cadmium uptake from the lumen of the rat jejunum (Foulkes, 1985) could be explained as the inhibition of rat Nramp2-dependent cadmium uptake by zinc. This discrepancy may be caused by species-specific differences in the transport functions of Nramp2 or by the difference of the Zn2⫹ transport study and the inhibition study of 109 Cd2⫹ uptake by Zn2⫹. Hg2⫹ (10 ␮M) also failed to inhibit 109Cd2⫹ uptake by human Nramp2. This result also corroborates the 55Fe uptake study using mouse intestine, which was not inhibited by 10 ␮M Hg2⫹ (Itturri and Nunez, 1998). These results implicated that the affinities of Zn2⫹ and Hg2⫹ for human Nramp2 are lower than that of Cd2⫹. Most of the cadmium in food products is usually bound to cadmium-binding proteins such as phytochelatin in cereal or metallothionein in meat. The gastrointestinal absorption of cadmium–phytochelatin (Fujita et al., 1993) and cadmium–metallothionein (Cherian et al., 1978; Sugawara and Sugawara, 1991) were studied and it was reported that these compounds are absorbed as complexes, which also enhances cadmium accumulation in the kidney. Although human Nramp2-mediated 0.1 ␮M cadmium uptake was inhibited by metallothionein (IC50 ⫽ 0.14 ␮M), lower concentrations of metallothionein failed to inhibit cadmium uptake in the present study. Since each metallothionein molecule is assumed to bind five to seven cadmium ions, the concentration of metallothionein in the lumen of the intestine is less than that of cadmium. Moreover, metallothionein may be partially degraded by digestion or gastric acidification, which would release free Cd2⫹ ion to be absorbed by human Nramp2. Since dietary iron was reported to lower the intestinal uptake of cadmium–metallothionein in rats (Groten et al., 1992), and the current studies demonstrate that human Nramp 2 is inhibited by Fe2⫹, then human Nramp2 appears to be the transporter for cadmium when cadmium– metallothionein is ingested orally. In conclusion, these studies are evidence that the human Nramp2 molecule functions as a zinc-insensitive, pH-dependent cadmium absorption transporter on the luminal membrane of human intestinal cells. Acknowledgments The authors thank Ms Akie Toki for her technical assistance. This research was supported by Comprehensive Research on Health Effects of Heavy Metals and Arsenic.

167

References Andrews, N.C., 2000. Intestinal iron absorption: current concepts circa 2000. Dig Liver Dis. 32, 56 – 61. Chen, X.Z., Peng, J.B., Cohen, A., Nelson, H., Nelson, N., Hediger, M.A., 1999. Yeast SMF1 mediates H(⫹)-coupled iron uptake with concomitant uncoupled cation currents. J. Biol. Chem. 274, 35089 –35094. Cherian, M.G., Goyer, R.A., Valberg, L.S., 1978. Gastrointestinal absorption and organ distribution of oral cadmium chloride and cadmiummetallothionein in mice. J. Toxicol. Environ. Health 4, 861– 866. Chow, C.W., Kapus, A., Romanek, R., Grinstein, S., 1997. NO3⫺-induced pH changes in mammalian cells: evidence for an NO3⫺–H⫹ cotransporter. J. Gen. Physiol. 110, 185–200. Elisma, F., Jumarie, C., 2001. Evidence for cadmium uptake through Nramp2: metal speciation studies with Caco-2 cells. Biochem. Biophys. Res. Commun. 285, 662– 668. Flanagan, P.R., McLellan, J.S., Haist, J., Cherian, G., Chamberlain, M.J., Valberg, L.S., 1978. Increased dietary cadmium absorption in mice and human subjects with iron deficiency. Gastroenterology 74, 841– 846. Foulkes, E.C., 1985. Interactions between metals in rat jejunum: implications on the nature of cadmium uptake. Toxicology 37, 117–125. Foulkes, E.C., 1991. Further findings on the mechanism of cadmium uptake by intestinal mucosal cells (step 1 of Cd absorption). Toxicology 70, 261–270. Fujita, Y., el Belbasi, H.I., Min, K.S., Onosaka, S., Okada, Y., Matsumoto, Y., Mutoh, N., Tanaka, K., 1993. Fate of cadmium bound to phytochelatin in rats. Res. Commun. Chem. Pathol. Pharmacol. 82, 357–365. Griffiths, W.J., Kelly, A.L., Smith, S.J., Cox, T.M., 2000. Localization of iron transport and regulatory proteins in human cells. QJM 93, 575–587. Groten, J.P., Luten, J.B., van Bladeren, P.J., 1992. Dietary iron lowers the intestinal uptake of cadmium–metallothionein in rats. Eur. J. Pharmacol. 228, 23–28. Gunshin, H., Mackenzie, B., Berger, U.V., Gunshin, Y., Romero, M.F., Boron, W.F., Nussberger, S., Gollan, J.L., Hediger, M.A., 1997. Cloning and characterization of a mammalian proton-coupled metal-ion transporter. Nature 388, 482– 488. Han, O., Fleet, J.C., Wood, R.J., 1999. Reciprocal regulation of HFE and Nramp2 gene expression by iron in human intestinal cells. J. Nutr. 129, 98–104. Iturri, S., Nunez, M.T., 1998. Effect of copper, cadmium, mercury, manganese and lead on Fe2⫹ and Fe3⫹ absorption in perfused mouse intestine. Digestion 59, 671– 675. Jarup, L., Berglund, M., Elinder, C.G., Nordberg, G., Vahter, M., 1998. Health effects of cadmium exposure: a review of the literature and a risk estimate. Scand. J. Work Environ. Health 24, 1–51. Ovesen, L., Bendtsen, F., Tage-Jensen, U., Pedersen, N.T., Gram, B.R., Rune, S.J., 1986. Intraluminal pH in the stomach, duodenum, and proximal jejunum in normal subjects and patients with exocrine pancreatic insufficiency. Gastroenterology 90, 958 –962. Phillpotts, C.J., 1979. Retention of cadmium in the duodenum of the rat following oral administration. Toxicology 14, 245–253. Picard, V., Govoni, G., Jabado, N., Gros, P., 2000. Nramp 2 (DCT1/ DMT1) expressed at the plasma membrane transports iron and other divalent cations into a calcein-accessible cytoplasmic pool. J. Biol. Chem. 275, 35738 –35745. Sorensen, J.A., Nielsen, J.B., Andersen, O., 1993. Identification of the gastrointestinal absorption site for cadmium chloride in vivo. Pharmacol. Toxicol. 73, 169 –173. Sugawara, N., Sugawara, C., 1991. Gastrointestinal absorption of Cd– metallothionein and cadmium chloride in mice. Arch. Toxicol. 65, 689 – 692. Sullivan, M.F., Ruemmler, P.S., 1987. Effect of excess Fe on Cd or Pb absorption by rats. J. Toxicol. Environ Health 22, 131–139.