Available online at www.sciencedirect.com

ScienceDirect

Journal of Nutritional Biochemistry 24 (2013) 1697–1708

Characterization of the GufA subfamily member SLC39A11/Zip11 as a zinc transporter

Yu Yua,b,1, Aimin Wua,b, c,1, Zhuzhen Zhanga, Guang Yanb, Fan Zhangb, Lihong Zhanga, Xiaoyun Shena,Ronggui Hud, Yan Zhangb,⁎, Keying Zhangc,⁎, Fudi Wanga,⁎

aDepartment of Nutrition, Institute of Nutrition and Food Safety, School of Public Health, Zhejiang University, Hangzhou 310058, P.R. ChinabKey Laboratory of Nutrition and Metabolism, Shanghai Key Laboratory of Pediatric Gastroenterology and Nutrition, Institute for Nutritional Sciences, Shanghai Institutes for Biological

Sciences, Chinese Academy of Sciences, Shanghai 200031, P.R. ChinacInstitute of Animal Nutrition, Sichuan Agricultural University, Sichuan Ya'an 625014, P.R. China

dInstitute of Biochemistry and Cell Biology, Shanghai 200031, P.R. China

Received 22 June 2012; received in revised form 29 January 2013; accepted 14 February 2013

Abstract

Cellular zinc influx and efflux are maintained by two major transporter families, the ZIP (SLC39A) and ZnT (SLC30A or CDF) molecules. The functions of onemolecule in this class, ZIP11/SLC39A11, remain unclear. Bioinformatics analysis of the distribution and evolutionary relationships of different ZIP members ineukaryotes and prokaryotes indicated that Zip11, the sole member of gufA subfamily, is an ancient ZIP family member that might have originated in earlyeukaryotic ancestors. Murine Zip11 mRNA is abundantly expressed in testes and the digestive system including stomach, ileum and cecum. Analysis of cellularzinc content, metallothionein levels, and cell viability under high or low zinc conditions in cells transfected with a murine Zip11 expression plasmid, suggest thatZip11 is a zinc importer. Further, cellular zinc concentrations and metallothionein levels decreased when Zip11 was knocked down. In mice supplemented withzinc, both mRNA and protein levels of Zip11 were slightly up-regulated in several tissues. The metal response element sequences (MREs) upstream of the firstexon of Zip11 responded to elevated extracellular zinc concentrations, as assessed by luciferase reporter assays. Mutagenic analysis showed that several of theMREs could regulate Zip11 promoter activity, and metal-responsive transcription factor-1 (MTF-1) was shown to be involved in this process. Collectively, thesedata suggest that Zip11 has unique protein sequence and structure features, it functions as a cellular zinc transporter, and its expression is at least partiallyregulated by zinc via hMTF-1 binding to MREs of the Zip11 promoter.© 2013 Elsevier Inc. All rights reserved.

Keywords: Zinc transporter; SLC39A11/ZIP11; ZIP family; MRE; MTF-1; Zinc homeostasis

1. Introduction

Zinc (Zn) is a trace element essential for life. More than 300metalloenzymes of six major functional classes require zinc as a keystructural component or as a cofactor. Moreover, zinc plays importantroles in DNA synthesis, cellular signal recognition, and secondmessenger metabolism. Zinc deficiency impairs growth, immuneactivity and brain function; however, zinc can also be toxic if over-accumulated. Thus, sophisticated regulatory systems must exist tomaintain zinc homeostasis at both the cellular and organismal levels

⁎ Corresponding authors. F. Wang is to be contacted at: Department ofNutrition, Institute of Nutrition and Food Safety, School of Public Health,Zhejiang University, Hangzhou 310058, P.R. China. K. Zhang, Institute ofAnimal Nutrition, Sichuan Agricultural University, Sichuan Ya'an 625014, P.R.China. Y. Zhang, Institute for Nutritional Sciences, Shanghai Institutes forBiological Sciences, Chinese Academy of Sciences, Shanghai 200031, P.R.China. Tel.: +86 571 88208582; fax: +86 571 88208099.

E-mail addresses: yanzhang01@sibs.ac.cn (Y. Zhang), zkeying@yahoo.com(K. Zhang), fwang@zju.edu.cn, fudiwang.lab@gmail.com (F. Wang).

1 Yu Yu and Aimin Wu contributed equally to this work.

0955-2863/$ - see front matter © 2013 Elsevier Inc. All rights reserved.http://dx.doi.org/10.1016/j.jnutbio.2013.02.010

[1–3]. In mammals, zinc balance is primarily regulated throughintestinal absorption, renal reabsorption, fecal elimination of excesszinc and the loss of endogenous zinc in the intestine through bothpancreatic and liver excretion.

Movement of zinc into and out of cells and subcellular organelles ismediated by zinc transporters [4]. Zinc transporters are largelyassigned to two metal-transporter families: ZIP (ZRT/IRT-like protein,SLC39A) and ZnT (Cation Diffusion Facilitator, or SLC30A). Theprevailing view is that members of the ZnT family transport zincfromwithin the cell, either out of the cell across the plasmamembraneor into intracellular compartments, which reduces cytosolic concen-trations. ZIP family members are believed to function in the oppositedirection, by increasing cytosolic zinc concentrations via uptake acrossthe plasma membrane or efflux from intracellular compartments[5–7]. The mammalian ZIP family consists of 14 members in 4subfamilies: ZIP subfamily I (ZIP9), ZIP subfamily II (ZIP1, ZIP2 andZIP3), LIV-1 family (ZIP4, ZIP5, ZIP6, ZIP7, ZIP8, ZIP10, ZIP12, ZIP13 andZIP14) and gufA family (ZIP11). Previously, ZIP1-8, ZIP10, and ZIP14have been shown to facilitate zinc uptake in cells [5,8–10].

Many reports document the regulation of gene expression by zinc,and identify zinc transporter genes as among those regulated, both

1698 Y. Yu et al. / Journal of Nutritional Biochemistry 24 (2013) 1697–1708

positively (e.g., ZnT1) and negatively (e.g., Zip4), by increased zincavailability [11–13]. Transcriptional regulation of gene expression byzinc occurs through binding of the Metal-responsive transcriptionfactor-1 (MTF-1), which is a six zinc-finger (Cys2His2) transcriptionfactor, that functions as a sensor of intracellular zinc by binding tometal response elements (MREs). The presence of MREs in genepromoter regions is well defined for mouse metallothionein (MT) Iand II genes, mouse ZnT1, ZnT2 and ZnT5 genes, and zebrafish Zip10gene [14–20].

Currently, the functions of Zip11, which is the sole member of thegufA subfamily of ZIP proteins, remain unknown. Here, we investi-gated both the regulation of Zip11 gene expression by zinc and thecellular functions of the Zip11 protein. We identified several MREsequences upstream of the first exon of Zip11, which appear to beinvolved in responses to elevated extracellular zinc concentrations.Moreover, through overexpression and knockdown studies we foundevidence that Zip11 is a zinc transporter that increases intracellularzinc content and MT mRNA expression.

2. Materials and methods

2.1. Bioinformatics analysis of ZIP family

Twenty-five representative eukaryotic genomes from different eukaryotic taxa(mammals, vertebrates, insects, nematodes, fungi, plants and protozoa) withsubstantial sequence coverage generated by the Entrez Genome Project at the NCBIwere used to analyze the distribution of organisms that express ZIP proteins. Foranalysis of the differential distribution of ZIP family members, the human ZIP1-14protein sequences were used as initial queries to search for homologous sequences inthese genomes using BLASTP and TBLASTN [21]. Orthologous proteins were defined onthe basis of the bidirectional best hit test [22] and phylogenetic analysis. Represen-tative archaeal and bacterial ZupT/gufA proteins (ZIP homologs) were also included.Multiple sequence alignments were performed using CLUSTALW [23] with defaultparameters. The phylogenetic tree was reconstructed with PHYLIP programs [24] andvisualized with PhyloDraw [25].

2.2. Animal experiments

All experiments involving mice were approved by the Institutional Animal Careand Use Committee of the Institute for Nutritional Sciences, Shanghai Institutes forBiological Sciences and Chinese Academy of Sciences. Male C57BL/6 mice (28 days old)were purchased from Shanghai Laboratory Animal Center of CAS and fed a standardrodent laboratory diet (Shanghai Laboratory Animal Center of CAS) for 1 week. Then,these mice were fed an AIN76-based diet (Research Diets, Inc.) containing 50 mg Zn/kgdiet (Zn-adequate, control) for 3 weeks. To examine the tissue expression of Zip11, sixmice were killed, and the indicated tissues were harvested and put in TRIZOL forsubsequent extraction of RNA and real-time PCR analysis. To examine the effects of Znon Zip11 expression, 8-week-old mice (six mice per group) were administered ZnSO4

(35 mg Zn/kg body weight) intragastrically and euthanized 3 hours (Zn-3h) or 8 hours(Zn-8h) later. The control group was administered the same volume of saline asexperimental mice. Control, Zn-3h, and Zn-8h groups were used in the Znsupplemental study. The indicated tissues were harvested and put in TRIZOL orsnap-frozen in liquid nitrogen for real-time PCR analysis and western blot. Plasmasamples were collected for assaying zinc content.

2.3. Cell culture

Human embryonic kidney 293T (HEK293T) cells, Raw 264.7 cells andMadin-Darbycanine kidney (MDCK) cells were cultured in Dulbecco's modified Eagle's medium(Invitrogen) plus 0.45% glucose under 5% CO2. All culturemedia contained 100 units/mlpenicillin, 100 g/ml streptomycin, 2 mM L-glutamine, and 100 μM non-essential aminoacids (Invitrogen) supplemented with 10% fetal bovine serum (PAA, Germany). Stablytransfected MDCK cell lines were selected with 500 μg/ml G418 (Sigma) for 2 weeksafter mZip11 transfection.

2.4. Plasmids constructs and site-directed mutagenesis

The open reading frame (ORF) of hZIP5, mZip11, or mZip14 were cloned into thepCMV-3tag-3A vector (Stratagene). Plasmids were confirmed by sequencingand protein expression by western blot.Mouse Zip11 promoter sequence was predictedusing UCSC Genome Bioinformatics and online software (Promoter 2.0 PredictionServer, http://www.cbs.dtu.dk/services/Promoter/). The Zip11 promoter construct(−2.04 kb to +278 bp relative to the start site of exon 1 of Zip11) was cloned fromgenomic DNA of C57BL/6 mouse, using PCR with primers containing the EcoR Irestriction site. The PCR product was then subcloned into the pGemT–Easy plasmid

using T/A cloning and further inserted into the pGL3 Basic plasmid (Promega) by usingSac I and Nhe I restriction sites. Plasmid constructs were confirmed by DNA sequencing.Then the other two Zip11 promoters (−305 bp or −123 bp to +278 bp relative to thestart site of exon 1 of Zip11) were cloned from the first promoter and inserted into thepGL3 Basic plasmid. The Zip11 Luc mutant constructs containing the 583 bp promoter(−305 bp to +278 bp) were created using site-directed mutagenesis (Quick Change II,Stratagene). Further confirmation of mutations and integrity of promoter fragmentswas performed by DNA sequencing. MRE sequences were identified via the Genomatixand USCS Genome Bioinformatic databases. The oligonucleotides used to generatethe mZip11 plasmid, three promoters, and mutations of the MREs sites are listedin Supplementary Table 1. The analysis of the promoter activity and the MRE mutationused in this experiment were the same as a previously report [26]. Endotoxin-free preparations of these plasmids (EndoFree Plasmid Maxi kit, Qiagen) were usedfor transfections.

2.5. RNA extraction and quantitative real-Time PCR (qRT-PCR)

HEK293T cells were grown in 6-well plates and transiently transfected with emptyvector or plasmids expressing hZIP5, mZip11, or mZip14 using FuGene 6 (Roch) for36 h. Transfections were performed following the manufacturer's instructions. Mediawere replaced with 2 ml of serum-free medium containing 100 μM Zn (ZnCl2) for 24 h.

Raw264.7 cells were grown in 12-well plates and transiently transfected withnegative control and mZip11 siRNA (Dharmacon, Thermo) using DharmaFECT siRNAtransfection reagents (Thermo) for 24 h. Transfections were performed following themanufacturer's instructions. Media was replaced with serum-free medium containing0 or 100 μM Zn (ZnCl2) for 24 h.

Total RNA was isolated from tissues or cells using Trizol (Invitrogen) according tothe manufacturer's instructions and treated with DNase I (Promega). RNA concentra-tion and purity were assessed by spectrophotometry. RNA (2.0 μg) was reverse-transcribed with M-MLV reverse transcriptase (Promega) and oligo (dT) 18 primers(Takara) as recommended. qRT-PCR was performed using a CFX96™ Real-Time System(Bio-Rad) and iQ™ SYBR Green Supermix (Bio-Rad) as described by the manufacturer.Raw data were normalized to the internal control, GAPDH, and presented as relativeexpression level calculated by 2ΔΔCt method. All primers for qRT-PCR are described inSupplementary Table 2. All experiments were performed in triplicate.

2.6. Western blot and immunofluorescence

HEK293T cells were grown in 6-well plates and transiently transfected with emptyvector or plasmids expressing hZIP5-Flag, mZip11-Flag, and mZip14-Flag for 48 h. Cellextracts were then prepared for western blot. Raw264.7 or MDCK cell lyses were usedto detect mZip11 protein expression.

Samples from lysed cells or tissues (30 μg total protein) were resolved on 12% SDS-PAGE gels, transferred to polyvinylidene fluoride membranes and probed with rabbitanti-mouse-Zip11 (1:200 dilution, Abcam), rabbit anti-Flag (1:1000 dilution, CellSignaling) or monoclonal antibody anti-β-actin (1:2000 dilution, Sigma), followed byeither anti-rabbit or anti-mouse IgG secondary antibodies conjugated to horseradishperoxidase at a dilution of 1:4000 (Protein Tech Group, Inc) and detection with the ECLSystem (Pierce).

HEK293T cells were grown in 24-well plates and transiently transfected with theempty vector, mZip11-Flag, or mZip14-Flag plasmid. After 36 h, the transfected cellswere incubated with 5 μM FluoZin-3AM (Molecular Probes, Gibco), a cell-permeablezinc fluorophore, in medium without FBS for 30 min. The cells were then stimulatedwith Zn2+ (80 μM) to measure intracellular Zn2+ accumulation by fluorescencemicroscopy. Raw264.7 cells were transiently transfected with negative control ormZip11 siRNA for 24 h, then treated as HEK293T cells to measure zinc fluorescence.

2.7. Zinc measurement with inductively-coupled plasma mass spectrometry (ICP-MS)

The study of using ICP-MS to measure zinc content in cells has been reportedpreviously [27]. HEK293T cells were grown in six-well plates and transientlytransfected with the empty vector, mZip11-Flag or mZip14-Flag plasmids for 36 h.Then the cells were treated with or without 100 μM Zn for 24 h then scraped off platesusing 5 ml of PBS ultra pure (137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 2 mMKH2PO4). Solutions were centrifuged and the cell precipitate was prepared for ICP-MSdetection. Protein was extracted from duplicate samples and measured by Bradford inSpecTRA MAX 190 microplate reader (Sunnyvale). Raw264.7 cells were grown in 6-well plates and transiently transfected following the protocol described above. MDCKcells stably expressed mZip11 or empty vector for 24 h, then media were replaced withserum-free medium containing 100 μMZn (ZnCl2), Cu (CuCl2), Fe (FAC), Mg (MgCl2) orMn (MnCl2) for 24 h. Cells were then prepared for ICP-MS and protein detection. Thepre-treatment method of cell and plasma for ICP-MS detection has been describedpreviously [28].

For analysis of metal concentrations, ICP-MS was performed using Agilent 7500cxICP/MS system (Agilent Technologies) equipped with a G3160B I-AS integratedautosampler. The G3148B ISIS system was used to reduce the detection time andvolume of each sample. Ni sample cone and skimmer cone were used with an orificediameter of 1.0 and 0.4 mm, respectively. Sample introduction was performed with amicromist nebulizer combined with a Scott-type double pass spray chamber (Agilent

1699Y. Yu et al. / Journal of Nutritional Biochemistry 24 (2013) 1697–1708

Technologies). The determination was operated in full quantitative mode, and theoperating conditions used in this study are the same as Zhao et al. [29]. The internalstandard was the multi-element standard solution (SPEX CertiPrep), injected byperistaltic pump into the ion source at an approximate concentration of 50 μg/L in theonline mode. Multi-element standard solutions of 24 elements, each at a concentrationof 1000 μg/ml (SPEX CertiPrep), were used, and zinc content was detected. Theconcentrations of the standard curve were 0, 2, 10, 50 and 250 μg/L.

2.8. Cell viability assays

HEK293T cells were seeded into 96-well plates and transfected with empty vector,hZIP5-Flag, or mZip11-Flag. Culture medium was removed 36 h after seeding andreplaced with 100 μl serum-free media containing the indicated concentration of TPENor ZnCl2. Cells were cultured for a further 48 hours, then 10 μl of CCK-8 Assay Kitreagent (Dojindo Molecular Technologies, Inc.) was added to each well. Plates wereincubated at 37°C for 2 h, and then absorbance was measured at 450 nm using SpecTRAMAX 190 microplate reader (Sunnyvale). The reference wavelength was 600 nm.Background absorbance (from wells without cells) was subtracted from all values.

2.9. Luciferase reporter assay

HEK293T cells were seeded into 24-well plates at 3.5×105 cells/well andtransfected with three promoters and Zip11 Luc mutant constructs. Some cells werecotransfected with hMTF-1 cDNA (Origene) to study whether MTF-1 was involved inregulation of the Zip11 transcript. All cells were cotransfected with the Renillaluciferase control plasmid pRL-SV40 (Promega) at the ratio of 1:200. Transfectionmedium was removed after 36 h and replaced with serum-free medium with orwithout 100 μM Zn for 24 h. Luciferase activity was measured with a Dual-LuciferaseAssay System (Promega) with a SpectraMax M5 (Molecular Devices) in luminometermode following the manufacturer's protocol. The raw values of firefly luciferase werenormalized to Renilla luciferase that had been transfected concurrently to correct fordifferences in transfection efficiency. The Renilla luciferase was stable throughout theexperiments, and not significantly affected by zinc status.

2.10. Electrophoretic mobility shift assay

EMSA was performed as described previously [30]. Cell extracts were preparedfrom HEK293T cells transfected with hMTF1 for 24 h and incubated with 100 μM Zn for30 min using NE-PER® Nuclear and Cytoplasmic Extraction Reagents (Thermo). Thebinding reactions were assembled and contained the following labeled, double-stranded oligonucleotides: MREs, a consensus MRE oligonucleotide that has a specific,high affinity MTF-1 binding site; MRE1-MRE5 are in the mouse Zip11 promoter. Oligosequences are listed in Supplementary Table 3. After 15 min at 4°C, the reactions werelysed and subjected to 4% PAGE. The MTF-1-MRE complexes were quantitated byautoradiography. Competition EMSAs were performed in which labeled MRE1-MRE5was incubated with MTF-1 in the presence of 200-fold molar excess of unlabeledcompetitor (MREs) in order to determine specificity following the manufacturer'sprotocol (LightShift® Chemiluminescent EMSA Kit, Pierce).

2.11. Statistical analyses

Results are expressed as means±S.E.M. Significant changes betweengroups were determined by an unpaired two-tailed Student's t test. Pb.05 wasconsidered significant.

3. Results

3.1. Occurrence and phylogeny of the ZIP family, and sequence analysisof ZIP11 proteins of different species

We examined the distribution and evolutionary relationships ofdifferent ZIP family members in eukaryotes (Fig. 1A). Among allmembers of the ZIP family, Zip2 (ZIP subfamily II) was only present inmammalian genomes, suggesting that it might be the most recentlyevolved ZIP member. On the other hand, orthologs of Zip11 (ZIPsubfamily gufA) were detected in both multicellular and a fewunicellular eukaryotes (such as Stramenopiles), suggesting that thisprotein might have arisen in a eukaryotic ancestor. Zip9 (ZIPsubfamily I), Zip7 and Zip13 (ZIP subfamily LIV-1) were detected inboth vertebrates and invertebrates, implying that these ZIP membersmay have first appeared in the metazoan ancestors. Zip1, Zip3, andZip6 were observed only in vertebrates and sea urchins, but not ininsects or nematodes. Other ZIP family members (Zip4, Zip5, Zip8,Zip10, Zip12, and Zip14) appear to have evolved after the vertebrate

split. In a few cases, orthologs of some ZIP members were notdetected (e.g., Zip7 in Ornithorhynchus anatinus and Zip3 in Daniorerio, probably because of incomplete sequence data). Interestingly,all bird species examined appeared to have lost several ZIP genesincluding Zip1, Zip4, and Zip5.

ZIP-like proteins were also found in fungi, plants, protozoa andprokaryotes. Phylogenetic analyses showed that the majority of ZIP-like proteins detected in unicellular eukaryotes and plants wereclustered with ZIP subfamily II (Zip1, Zip2 and Zip3 members),including the ZRT proteins in fungi and IRT proteins in plants, bothknown to be involved in zinc import in these organisms (Fig. 1B). Thus,it appeared that these ZIP-like proteinsmight have evolved before thesplit of ZIP1/2/3 members, implying that the ancestor of the ZIPsubfamily II was present in early eukaryotes. Further analysis of theprokaryotic ZIP homologs (ZupT/gufA proteins) revealed that theywere clustered with Zip11 proteins (Fig. 1B), strongly suggesting thatZip11/ZupT/gufA family is themost ancient zinc transporter present inthe last universal common ancestor (LUCA). It is possible that manyunicellular organisms and plants have lost this gene whereasmetazoangenomes still contain this gene, albeit, for unknown reasons.

Most mammalian ZIP proteins, including Zip11, are predicted tohave eight transmembrane domains. Multiple sequence alignment ofZip11 and its homolog, ZupT/gufA, revealed that these transmem-brane domains are conserved in both prokaryotes and eukaryotes(Supplementary Figure 1). The accessionnumbers for Zip11 sequencesare listed in Supplementary Table 4. In comparison to other ZIPproteins, Zip11 had very few histidines (His) and almost no cysteine(Cys) residues. For example, human andmouse Zip11 have only threeHis residues and no Cys residues were found. We hypothesized thatthe last His, which exists in transmembrane domain 4 and wasconserved in all organisms examined, might be involved in the zinc-binding process. Thus, suggesting that the process by which Zip11binds to zinc might be different from other ZIP members.

3.2. Murine Zip11 mRNA expression pattern in tissues

Themurine Zip11 (mZip11) tissue expression patternwas assessedby Real-Time PCR (Fig. 2A). We found thatmZip11 is highly expressedin the testes and portions of the digestive system including thestomach, ileum and cecum. In contrast,mZip11was expressed at verylow levels in liver, duodenum, jejunum, and colon. Zip4, the major gutzinc importer, was highly expressed in the digestive system, includingthe duodenum, jejunum and ileum, and colon (Fig. 2B). In general, adirect comparison of mRNA levels showedmZip11was expressed at amuch lower level than mZip4, except in stomach and testes (Fig. 2Aand 2B). This expression pattern suggests that Zip11 may haveimportant functions in the upper digestive system as well as the malereproductive organs.

3.3. Murine Zip11 is capable of cellular zinc uptake

To gain an appreciation of the role of Zip11 in zinc trafficking, weassessed the effect of Zip11 overexpression.We first checkedmZip11-Flag protein expression by western blot, using human Zip5 (hZip5)and murine Zip14 (mZip14) as controls (Fig. 3A). The mZip11-Flagprotein was detected as a ~37 kDα band as predicted; however, anadditional ~74 kDα band was also detected, possibly representing amZip11 dimer or non-specific band, and it needs to be furtheridentified. Both hZip5-Flag and mZip14-Flag proteins were detectedat their predicted molecular weights. Zinc uptake was monitored viageneration of fluorescence following interaction of intracellular, labileZn2+with FluoZin-3AM. Four hours after adding zinc at 80 μM, intensefluorescence was detected in cells transfected with mZip11-Flag orthe positive control, mZip14-Flag, whereas virtually no fluorescencewas detected in cells transfected with the empty vector alone

Fig. 1. Bioinformatics analyses of the ZIP family. (A) Occurrence of different members of the ZIP family in eukaryotes. A set of model organisms (from unicellular organisms to humans)was selected to illustrate the distribution of Zip1-Zip14 members in eukaryotes. In each organism the presence of each ZIP member is indicated. The phylogeny of four ZIP subfamiliesand ZIP members are also shown. (B) A phylogenetic tree of ZIP proteins in eukaryotes. A total of 130 ZIP proteins from representative organisms were used. Names of ZIP subfamiliesand ZIPmembers are shown in blue and red, respectively. Proteins that belong to each ZIPmember are shaded. The branches separating archaeal and bacterial ZupT/gufA proteins havebeen shortened for illustration purposes.

1700 Y. Yu et al. / Journal of Nutritional Biochemistry 24 (2013) 1697–1708

Fig. 2. Expression patterns of Zip11 and Zip4mRNAs in mouse tissues. The mRNA abundance of Zip11 (A) and Zip4 (B) in mouse tissues was assayed by qRT-PCR. Six, 8-week-old maleC57BL/6 mice were used in the experiments. Results were normalized to the GAPDH internal control and are presented as relative expression level calculated by the 2ΔΔCt method. Alldata were normalized to the Zip4 mRNA level in brain, and expressed as means±S.E.M.

1701Y. Yu et al. / Journal of Nutritional Biochemistry 24 (2013) 1697–1708

(Fig. 3B). We also used ICP-MS to measure zinc content in transfectedcells. Zinc concentrations increased slightly, but significantly in cellstransfected with mZip11-Flag or mZip14-Flag compared to cellstransfected with empty vector alone (Fig. 3C, gray bars). Additionally,the zinc content of cells overexpressing mZip11-Flag or mZip14 withsupplemental zinc was much higher than that of cells transfectedwith empty vector (Fig. 3C, green bars), also indicating that Zip11 actsas a zinc importer. Further evidence that Zn2+ uptake was occurringthrough Zip11 was the increased MT mRNA levels in transfected cells(Fig. 3D). Lastly, empty vector, mZip11-Flag transfected, and hZIP5-Flag-transfected [8] cells were incubated with increasing concen-trations of zinc and a zinc chelator, TPEN (N,N,N′N′-tetrakis (−)[2-pyridylmethyl]-ethylenediamine), for 2 days, and cell viabilitieswere measured. We found that viability of cells transfected with

mZip11-Flag or hZIP5-Flag was lower in high zinc supplementedmedium (Fig. 3E) and higher in TPEN added medium (Fig. 3F) thanthat in cells transfected with empty vector. The mZip11-Flag andhZIP5-Flag transfected cells had increased sensitivities to high zinc(Fig. 3E) and resistance to the chelator TPEN (Fig. 3F). These resultswere consistent with the ability of mZip11 to transport zinc into thecell where it is then available for zinc-dependent processes.

In addition, mZip11 knockdown with siRNA in Raw264.7 cellsconfirmed a role for mZip11 in zinc uptake. Expression of mZip11 wasreduced after siRNA treatment of cells (Fig. 4A and 4B). By using thezinc indicator, FluoZin-3AM, we were able to detect a decrease in zinc-associated fluorescence when mZip11 expression was knocked down(Fig. 4C). We also used ICP-MS to measure the zinc content oftransfected cells. Zinc concentrations were significantly reduced in

1702 Y. Yu et al. / Journal of Nutritional Biochemistry 24 (2013) 1697–1708

1703Y. Yu et al. / Journal of Nutritional Biochemistry 24 (2013) 1697–1708

cells transfected with mZip11 siRNA compared to cells transfectedwith negative control (Fig. 4D). Furthermore, mZip11 knockdownmarkedly reduced MT expression (Fig. 4E).

To assess the substrate specificity of mZip11, we examined theeffects of various metals on ion accumulation by MDCK cells stablyexpressing mZip11-Flag. We found mZip11 appeared to also beinvolved in zinc and copper intracellular trafficking (Fig. 5A and 5B). Itis known that other transporters are not metal specific, such as Zip14,which can transport both iron and zinc [9,31].

3.4. The presence of available zinc up-regulates murine Zip11 bothtranscriptionally and translationally in vivo

In order to investigate whether Zip11 expression was regulated byzinc, we gavage-fed mice with a dose of 35 μg Zn/g body weight orsaline (control) and euthanized them three (Zn-3h) or eight hours(Zn-8h) later. First, zinc concentrations in plasma were analyzedusing the ICP-MS detection system. Plasma zinc levels increasedapproximately two-fold after eight hours of zinc repletion (Supple-mentary Figure 2). We then measured both MT and Zip11 mRNAlevels in stomach, testes, liver, and spleen (Fig. 6A), normalizing datain the treatment groups to the control group. Transcript levels ofmouse MT increased with the addition of supplemental zinc in theliver and spleen. In contrast, Zip11 mRNA levels in stomach, testes,and liver did not change with zinc supplement, but splenic levelsincreased approximately two-fold after eight hours of zinc repletion(Fig. 6A). To extend these analyses, Zip11 protein levels were assessedby western blot analysis, and Zip11/actin ratios were quantitated bydensitometry for three independent experiments. Zip11 proteinexpression was found to be significantly increased after Zn addition,with the difference in the liver and spleen being substantial (Fig. 6B).

3.5. MTF-1 binding to the mZip11 MRE is required for enhancedgene expression

Ten MRE-like sequences are predicted in the promoter region ofZip11. To further investigate the potential roles of these MREs, wecloned a fragment spanning −2040 bp to +278 bp of the Zip11promoter, as well as smaller fragments from −305 bp or −123 bp to+278 bp, into the pGL3-Basic vector. The locations and sequences ofthe potential MREs for these three fragments, 2318 bp (P2318), 583bp (P583) and 401 bp (P401) are given in Fig. 7A and 7B. We foundthat the promoter activity of P583 and P401 were much higher inthe presence of 100 μM zinc than without addition of zinc (Fig. 7C,closed bars). These results suggest that the two small promoterfragments, P401 and P583, contain regions important for regulationof Zip11 mRNA expression by zinc. Subsequently, we used P583 as atemplate to generated MRE mutant constructs (Fig. 7B) because thebasal activity of P583 was the highest of three promoters (data notshown). We found that the promoter activities increased signifi-cantly in the presence of zinc only with MRE1, MRE2, or MRE3 andMRE5 together (Fig. 7D, open bars). Cells were co-transfected withthese mutant constructs and a human MTF-1 (hMTF-1) expressionplasmid to determine the importance of these MRE sequences and

Fig. 3. Murine Zip11 transports zinc into cells. (A) The Zip-Flag protein levels in the transfeexpressing hZIP5-Flag, mZip11-Flag, or mZip14-Flag. Total protein extracts were prepared anwith empty vector (Right), mZip11-Flag (Middle) or mZip14-Flag (Left) was measured with Flufor 4 h after Fluozin-3AM treating. (C) Zinc content of transiently transfected cells was assess(open bars) or with (green bars) 100 μM zinc for 24 h. Cell lysis solutions were used to detect zAnalysis of MT mRNA levels in cells. Cells were transiently transfected with plasmids for 36relative MT mRNA abundance. All data were normalized to the value in cells transfected withzinc supplement. Cells overexpressing the indicated proteins were grown in serum-free mediacell viability (n=6). (F) The viabilities of cells supplemented with TPEN. Cells overexpressing thconcentration of TPEN for 2 days prior to analyzing cell viability (n=6). Each point is represcontrol group, Pb.05. In Fig. 3E and 3F, asterisks indicate significant differences between cells

MTF-1 in the regulation of Zip11 promoter activity (Fig. 7D, closedbars). We found that promoter activities were elevated significantlyby zinc in all groups with MRE2, MRE3, MRE4, or MRE5. Further,these results indicate that MRE1 has no effect on promoter activitywhile MRE5 plays a positive role. Intriguingly, the promoteractivities of constructs without MRE4 were markedly higher thanthat of P583 suggesting that MRE4 exhibits a negative effect onpromoter activity. Together, these findings suggest that MTF-1 isinvolved in the regulation of Zip11 promoter activity. Subsequently,through EMSA we determined MTF-1 can bind to these MREsequences (Supplementary Figure 3A), and this binding wasdependent on an intact core sequence (TGCRCNC) [14]. However,the MRE2, MRE3, and MRE4 core sequence, TGCNCNC, differs fromthe consensus at the fourth base. Cross-competition experimentsdemonstrated that a 200-fold molar excess of unlabeled MREseliminated binding (Supplementary Figure 3B).

4. Discussion

By conducting a thorough computational analysis, we revealed thedistribution of ZIP family members and their homologs in eukaryotesand prokaryotes. Among them, Zip11, the prokaryotic ZupT/gufAhomolog and sole member of the gufA subfamily, may have evolvedfrom ancestral eukaryotes. However, it appears that the Zip11 genewas lost in almost all unicellular organisms and plants. With eighttransmembrane domains almost completely lacking in His and Cysresidues, the mechanism by which Zip11 binds zinc is likely differentfrom other ZIP family members.

ZIP genes are differentially expresses across tissues; therefore,individual Zip family members may possess unique functions. Forexample, abundant expression of hZIP4 was identified in tissuesinvolved in zinc absorption/reabsorption, such as the smallintestine, stomach, colon and kidney [32]. Human ZIP5 displays asimilar pattern of tissue-specific expression to both mouse andhuman ZIP4, with high expression levels in the liver, kidney,pancreas and throughout the small intestine and colon [33]. Zip8is expressed in lung, kidney, testes, liver, brain, small intestine andthe membrane fraction of mature RBCs [34,35]. Zip10 is foundabundantly expressed in tissues such as small intestine, pancreas,testes, brain, liver and kidney. In mice, Zip10 has the highestexpression in the liver and brain [10,15]. Additionally, in the C57BL/6J mouse, Zip14a expression is highest in liver, duodenum, kidneyand testes, whereas Zip14b expression is highest in liver, duodenum,brain and testes [5]. In this study, we confirmed that mZip4 is highlyexpressed in the small intestine (duodenum, jejunum and ileum) aswell as in the colon (Fig. 2B), as reported previously [33]. ThemZip11 mRNA levels were highest in the stomach and testes among19 tissues examined, with lower expression in other tissues(Fig. 2A). The expression pattern of Zip11 might reflect its role inzinc metabolism in these tissues. For example, zinc is believed tohave important functions in the reproductive system [4]. Futurestudies will be required to determine if Zip11 is important instomach zinc absorption, similar to the reported function of Zip4 inintestinal zinc absorption [11].

cted cells. HEK293T cells were transiently transfected with empty vector or plasmidsd analyzed by western blot. (B) Intracellular labile zinc in cells transiently transfectedozin-3AM. Cells were transfected with plasmids for 36 h, and then treated by 80 μM zinc

ed by ICP-MS. Cells were transfected with plasmids for 36 h, and then treated withoutinc content using the ICP-MS detection system. Three samples per group were used. (D)h, and then treated with 100 μM zinc for 24 h. Cell lysis solutions were used to detectempty vector alone. Four samples per group were analyzed. (E) Cell viability after highsupplemented with the indicated concentration of zinc for two days prior to analyzinge indicated proteins were grown in serum-freemedia supplemented with the indicatedented by means±S.E.M. Asterisks indicate differences between treated groups and thetransfected with mZip11-Flag and controls, Pb.05.

Fig. 4. MurineZip11 facilitates zinc uptake.Decrease of intracellular zinc concentrationaftermZip11knockdown. (A)RAW264.7 cellswere transfectedwithmZip11 siRNAor anegative control andincubated for 24 h.mZip11mRNAwas assessed by qRT-PCR. (B) The efficiency of mZip11 knockdownwas further analyzed bywestern blot with β-actin as a loading control. (C) RAW264.7 cellswere transfectedwith siRNA as above. Intracellular labile zinc was detected via the FluoZin-3AM fluorescence assay and is reported as relative fluorescence units detected by flow cytometry. (D)Following siRNA transfection, RAW264.7 cells were incubatedwith 100 μMzinc for 24 h, cells were lysed and zinc content assessed using the ICP-MS detection system. (E) RAW264.7 cells weretransfected with siRNA as above and 24 h later, incubated with 0 or 100 μM zinc for 24 h.MTmRNAwas measured by qRT-PCR. All data were normalized to the value in cells transfected withnegative control. Each point represents themean±S.E.M. Experiments were repeated in triplicate. Asterisks indicate significant differences between treated groups and the control group, Pb.05.

1704 Y. Yu et al. / Journal of Nutritional Biochemistry 24 (2013) 1697–1708

Intracellular zinc homeostasis is maintained by physiologicalprocesses including uptake, subcellular organelle sequestration andrestoration, and export. Members of the ZIP family have beendemonstrated to be involved in zinc uptake and in the release of

stored zinc into the cytoplasm of cells when zinc is deficient. Inmammals, Zip1-8, Zip10 and Zip14 proteins have been reported tofunction as zinc uptake proteins [5,8–10]. For instance, in transfectedcells, mZip5 functions in zinc uptake and is specific for zinc as a

Fig. 5. Zip11 is specific for zinc and copper. (A) mZip11-Flag protein level in stablytransfected cells. MDCK cells were transiently transfected with empty vector or stablytransfected with mZip11-Flag. Total protein extracts were prepared and analyzed bywestern blot. (B) Transfected MDCK cells were grown for 36 h and then treated withindicated metals at a concentration of 100 μM for 24 h. All competitor metals wereadded as divalent cations. Metal concentrations were measured after lysing cells usingthe ICP-MS detection system. Each point represents the mean of a representativeexperiment (n=3), and the error bars indicate ±1 SEM. Asterisks indicate significantdifferences between treated groups and the control group, Pb.05.

1705Y. Yu et al. / Journal of Nutritional Biochemistry 24 (2013) 1697–1708

substrate [8]. Transfection of Zip7 into cells causes an increase inintracellular zinc, as would be expected for a zinc importer [36]. Zip8functions as a divalent cation transporter for Mn, Zn and Cd in mousefetal fibroblast (MFF) cultures [37,38]. However, specificity fortransport of zinc could not be shown until studies of inhibition ofCd influx were conducted in Zip8 mRNA-injected Xenopus oocytes[34]. Our study demonstrates that Zip5 and Zip14 are zinc importers,in accordance with previous data [8,9]. Moreover, we show evidencethat Zip11 is a zinc importer (Figs. 3 and 4). Kelleher et al. showedZip11 localization in the Golgi apparatus [39]. Therefore, Zip11probably plays an important role in zinc trafficking in this organelle.Aydemir et al. found liver zinc content increased in response to Zip11mRNA upregulation after partial hepatectomy. This result suggestedZip11 might have function in facilitating zinc mobilization intohepatocytes similarly to Zip14, but this hypothesis will require furtherstudy [40]. The mechanism of transport used by Zip proteins is stillunresolved, largely because of conflicting results regarding theproperties of different transporters. For example, in yeast both Zrt1and Zrt2 are dependent on energy for zinc transport [41]. In contrast,neither hZIP1 [42] nor hZIP2 [43] requires ATP for activity.Additionally, electrogenic experiments in Xenopus oocytes revealedthat Zip8-mediated divalent cation movement across the membraneoccurs as Cd2+/[HCO3−]2 and Zn2+/[HCO3−]2 electroneutral com-plexes [34]. Expression of some zinc transporters appears to beregulated by post-transcriptional mechanisms, and may be depen-dent on zinc availability. Studies of transfected cells revealed that Zip1is mainly present in intracellular organelles in cells cultured in zinc-adequate medium, but is recruited to the cell surface when zinc is

limited, suggesting a posttranscriptional regulatory mechanism [44].During dietary zinc deficiency, Zip4 localizes to the apical membranesof enterocytes in the intestine and visceral endoderm cells in theembryonic visceral yolk sac [11,45]. However, zinc repletion wasreported to cause rapid endocytosis of Zip4 [46]. Zip5 is localized tothe basolateral surface of these cell types under zinc-repleteconditions but is internalized during periods of dietary zinc deficiency[12]. Thus, it remains unresolved whether Zip11 transports zinc inresponse to changes in energy, the presence of ion exchange channels,or through post-transcriptional regulation mechanisms. In addition,our data provided evidence that Zip11 probably transports copperinto cells in addition to zinc (Fig. 5). This non-specific metaltrafficking of Zip11 is similar to that of Zip8 and Zip14 [31,37]. Atpresent, the Cu/Zn ratio has been reported to be associated withmultiple abnormalities and diseases, indicating the tight couplingbetween zinc and copper [47].

Expression of many ZIP proteins has been shown to be regulatedby zinc. Dufner-Beattie et al. found thatmZip4mRNA and Zip4 proteinlevels in the small intestine increased significantly if dietary zinc wasrestricted, and decreased when dietary zinc was replete [12]. Inaddition, Zip4 mRNA and protein were reduced in Caco-2 cellscultured with high (200 μM) zinc compared with low (100 μM) zincconditions [48]. Lichten et al. analyzed ZIP familymembers in both thebrain and liver of zinc-deficient animals and found a marked increasein Zip10 expression, which then decreased in response to zincrepletion [15]. In addition, the protein abundance of Zip7 was foundto be repressed by supplemental zinc [5]. In contrast, expression ofmany other zinc transporters exhibits little response to zinc. Forexample, the abundance of Zip1 mRNA was not found to be regulatedby dietary zinc in the intestine or visceral endoderm, tissues involvedin nutrient absorption [49]. Additionally, Zip5 mRNA and proteinabundance are irresponsive to zinc [46]. In our study, we found thatZip11 mRNA in the stomach, liver or testes was not significantlyaffected by zinc, but in the spleen, Zip11 mRNA increased with zincrepletion (Fig. 6A). Further, Zip11 protein increased after Zn addition,with a marked difference in the liver and spleen (Fig. 6B). It is veryinteresting that increased Zip11 expression in response to increasedzinc availability is contrary to other Zip transporters, such as Zip4 orZip10. However, the regulation of Zip11 by zinc is similar to ZnT5 [17].The mechanism underlying this difference in expression betweenZip11 and other Zip transporters requires further study. Additionally,the regulation of Zip11 by zinc is tissue-specific. Similar findings havebeen reported for other zinc transporters. For example, in response toan acute oral Zn dose, the level of intestinal ZnT1 mRNA was up-regulated eightfold, without a corresponding increase in ZnT1protein. Conversely, the acute oral dose did not affect liver ZnT1mRNA, but resulted in a 5-fold increase in liver ZnT1 protein.Additionally, dietary zinc supplementation elevated the level ofintestinal ZnT1 mRNA and protein approximately 50% and 10%,respectively, but had no effect in the liver [50].

The mechanisms through which zinc regulates expression of zinctransporter genes have been studied extensively. By using RNA andprotein synthesis inhibitors and run-on transcription assays, in-creased expression of Zip4 during zinc deficiency was shown to becaused by stabilization of Zip4 mRNA, not increased transcriptionefficiency. Moreover, zinc repletion was reported to cause mRNAdegradation and rapid endocytosis of Zip4 [46]. Recently, however,the transcription factor Krüppel-like factor 4 (KLF4), which is inducedduring zinc restriction, was identified as a positive regulator of Zip4transcription [5]. In higher eukaryotes, the best-understood metal-regulated genes are within the MT family. Transcription of the mouseMT-I gene, for example, is regulated by zinc and cadmium, and thisregulation is mediated by MTF-1. MTF-1 is activated by zinc to bind toMREs in the MT-I promoter, resulting in an increased rate oftranscription [16,18,19]. Moreover, zinc-induced MTF-1 was found

Fig. 6. Expression of Zip11 in mouse tissues following supplementation with zinc. Mice were fed a zinc-adequate diet for 21 days, and were gavaged to supply a dose of 35 μg zinc/gbody weight or saline (control) and euthanized three (Zn-3h) or eight hours (Zn-8h) later. (A) TheMT and Zip11mRNA levels in stomach, testes, liver or spleen were analyzed by qRT-PCR. Values presented are relative mRNA normalized to GAPDH. The treatment group was normalized to the control group. (B) Zip11 protein levels in mice in stomach, testes, liver orspleen were analyzed by western blot, and Zip11/actin ratios were quantitated by densitometry for three independent experiments. All data are expressed as means±S.E.M., n=4–6.Asterisks indicate significant differences between treated groups and the control group, Pb.05.

1706 Y. Yu et al. / Journal of Nutritional Biochemistry 24 (2013) 1697–1708

to up-regulate the primary zinc exporter ZnT1 [14]. Guo et al. showedthat zinc regulation is conferred by the consensusMRE located at+53to +59 of the murine ZnT2 gene, which ablates zinc responsivenesswhen mutated [20]. The dramatic increase in expression of ZnT2withMTF-1 overexpression strongly supports a MTF-1/MRE mode ofregulation involving this MRE site. Additionally, Lichten et al. haveshown that MTF-1 can act as a repressor of Zip10 under normalcellular levels of labile zinc [15]. Upon reducing cellular zinc,repression is lost, as MTF-1 is not translocated to the nucleus, therebyallowing enhanced Zip10 transcriptional activation. The cause of thisapparent differential mode of MTF-1 action resides in the genomicplacement of the MRE downstream of the Zip10 transcriptional startsite. In our study, we found that many MRE sequences exist in thepromoter of Zip11, suggesting that MTF-1/MRE might also beimportant in Zip11 transcript regulation by zinc. We predicted theZip11 promoter sequence and made three promoter fragmentconstructs. Our results indicate that the promoter activities of P583and P401 are higher in the presence of 100 μM zinc than withoutaddition of zinc (Fig. 7C). Furthermore, the P583 and P401 fragmentsare important for upregulation of Zip11mRNA expression by zinc. Weused P583 as a template to generatedMREmutant constructs (Fig. 7B)based on the basal activity of P583 being highest of the threepromoters (data not shown).We found that whileMRE1 had no effecton promoter activity, MRE5 plays a positive role and MRE4 appears tofunction as a negative regulator. Additionally, MTF-1 was involved inthe regulation of Zip11 promoter activity (Fig. 7D), presumably via

binding to MRE sequences (Supplementary Figure 3). Moreover, thepresence of MRE4 may interfere with the binding of MTF-1 to MRE3or MRE5. Collectively, we conclude that the MTF-1/MRE reaction ispartially involved in the regulation of Zip11 transcript levels. Zip11mRNA in some mouse tissues is up-regulated with supplementalzinc treatment. Whether other mechanisms, including increasedmRNA transcript efficiency and the splicing and modification ofmRNA, have functions in the tissue specificity of Zip11 expressionremains to be determined.

In summary, our current study reveals critical functions of the ZIPfamily member, Zip11, which acts to transport zinc into thecytoplasm. Zip11 mRNA and protein levels increase with zincrepletion in some mouse tissues. In addition, MRE sequencesupstream of the first exon of Zip11 are involved in response toelevated extracellular zinc concentration. Of 19 tissues surveyed,Zip11 mRNA levels were highest in the stomach and testes. Byconducting a thorough computational analysis, we concluded thatZip11 is the sole member of the gufA subfamily of ZIP proteins, mayhave evolved from ancestral eukaryotes, and is lacking His and Cysresidues. Collectively, these data illustrate the unique role of Zip11 inzinc metabolism and suggest other physiological functions.

Acknowledgments

This work was supported by National Natural Science Foundationof China grants (No. 31225013 and 31030039 to F.W., 30901193 to

Fig. 7. Zip11 promoter activity. (A) Potential promoter regions and directions are listed. The start site of exon 1 is designated as +1. Each MRE motif in the promoter is marked by a blackarrow, indicating the direction of the MRE sequence. The locations and sequences of MREs are also given. (B) The locations, sequences, andmutations of the first fiveMRES. (C) The relativeluciferase activities of P2318, P583, and P401 following zinc supplementation. HEK293T cells were transfectedwith luciferase reporters and P2318, P583 and P401 constructs, incubated for24 h in serum-free media with (open bars) or without (closed bars) addition of 100 μM zinc supplement. Asterisks indicate a significant difference with P2318, Pb.05. (D) The relativeluciferase activities of P583 MRE mutants. HEK293T cells were cotransfected with MRE mutation constructs along with empty vector (open bars) or hMTF-1 (closed bars) for 36 h beforeaddition of 100 μMzinc for 24 h. The P583mutants with normalMREs are indicated at the bottom of the table. A significant difference between P583+Zn cells with corresponding P583 (nozinc, gray bar) is indicated by an asterisk, Pb.05. A significant difference between P583+hMTF-1+Zn cells with corresponding P583+ hMTF-1 (no zinc, black bar) is indicated by a #Pb.05.

1707Y. Yu et al. / Journal of Nutritional Biochemistry 24 (2013) 1697–1708

1708 Y. Yu et al. / Journal of Nutritional Biochemistry 24 (2013) 1697–1708

Y.Y., 81100943 to L.Z.); Shanghai Key Laboratory of PediatricGastroenterology and Nutrition (grant number 11DZ2260500);China Postdoctoral Science Foundation (No. 2012M510902 to Y.Y.)and SIBS-CAS Outstanding Youth Fellowship (No. 2010KIP309 toY.Y.). This work is also supported by Distinguished ProfessorshipProgram from Zhejiang University to F.W. We appreciate theencouragement and helpful comments from other members of theWang laboratory.

Appendix A. Supplementary Data

Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.jnutbio.2013.02.010.

References

[1] Vallee BL, Falchuk KH. The biochemical basis of zinc physiology. Physiol Rev 1993;73:79–118.

[2] Berg JM, Shi Y. The galvanization of biology: a growing appreciation for the rolesof zinc. Science 1996;271:1081–5.

[3] Koh JY, Suh SW, Gwag BJ, He YY, Hsu CY, Choi DW. The role of zinc in selectiveneuronal death after transient global cerebral ischemia. Science 1996;272:1013–6.

[4] Kambe T, Yamaguchi-Iwai Y, Sasaki R, Nagao M. Overview of mammalian zinctransporters. Cell Mol Life Sci 2004;61:49–68.

[5] Lichten LA, Cousins RJ. Mammalian zinc transporters: nutritional and physiologicregulation. Annu Rev Nutr 2009;29:153–76.

[6] Palmiter RD, Huang L. Efflux and compartmentalization of zinc by members of theSLC30 family of solute carriers. Pflugers Arch 2004;447:744–51.

[7] Liuzzi JP, Cousins RJ. Mammalian zinc transporters. Annu Rev Nutr 2004;24:151–72.

[8] Wang F, Kim BE, Petris MJ, Eide DJ. The mammalian Zip5 protein is a zinctransporter that localizes to the basolateral surface of polarized cells. J Biol Chem2004;279:51433–41.

[9] Liuzzi JP, Lichten LA, Rivera S, Blanchard RK, Aydemir TB, Knutson MD, et al.Interleukin-6 regulates the zinc transporter Zip14 in liver and contributes to thehypozincemia of the acute-phase response. Proc Natl Acad Sci USA 2005;102:6843–8.

[10] Kaler P, Prasad R. Molecular cloning and functional characterization of novel zinctransporter rZip10 (Slc39a10) involved in zinc uptake across rat renal brush-border membrane. Am J Physiol Renal Physiol 2007;292:F217–29.

[11] Dufner-Beattie J, Wang F, Kuo YM, Gitschier J, Eide D, Andrews GK. Theacrodermatitis enteropathica gene ZIP4 encodes a tissue-specific, zinc-regulatedzinc transporter in mice. J Biol Chem 2003;278:33474–81.

[12] Dufner-Beattie J, Kuo YM, Gitschier J, Andrews GK. The adaptive response todietary zinc in mice involves the differential cellular localization and zincregulation of the zinc transporters ZIP4 and ZIP5. J Biol Chem 2004;279:49082–90.

[13] Liuzzi JP, Blanchard RK, Cousins RJ. Differential regulation of zinc transporter 1, 2,and 4 mRNA expression by dietary zinc in rats. J Nutr 2001;131:46–52.

[14] Langmade SJ, Ravindra R, Daniels PJ, Andrews GK. The transcription factor MTF-1mediates metal regulation of the mouse ZnT1 gene. J Biol Chem 2000;275:34803–9.

[15] Lichten LA, Ryu MS, Guo L, Embury J, Cousins RJ. MTF-1-mediated repression ofthe zinc transporter Zip10 is alleviated by zinc restriction. PLoS One 2011;6:e21526.

[16] Wimmer U, Wang Y, Georgiev O, Schaffner W. Two major branches of anti-cadmium defense in themouse: MTF-1/metallothioneins and glutathione. NucleicAcids Res 2005;33:5715–27.

[17] Jackson KA, Helston RM, McKay JA, O'Neill ED, Mathers JC, Ford D. Splice variantsof the human zinc transporter ZnT5 (SLC30A5) are differentially localized andregulated by zinc through transcription and mRNA stability. J Biol Chem 2007;282:10423–31.

[18] Koizumi S, Suzuki K, Ogra Y, Yamada H, Otsuka F. Transcriptional activity andregulatory protein binding of metal-responsive elements of the human metal-lothionein-IIA gene. Eur J Biochem 1999;259:635–42.

[19] Daniels PJ, Andrews GK. Dynamics of the metal-dependent transcription factorcomplex in vivo at the mouse metallothionein-I promoter. Nucleic Acids Res2003;31:6710–21.

[20] Guo L, Lichten LA, Ryu MS, Liuzzi JP, Wang F, Cousins RJ. STAT5-glucocorticoidreceptor interaction and MTF-1 regulate the expression of ZnT2 (Slc30a2) inpancreatic acinar cells. Proc Natl Acad Sci USA 2010;107:2818–23.

[21] Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. Basic local alignment searchtool. J Mol Biol 1990;215:403–10.

[22] Tatusov RL, Galperin MY, Natale DA, Koonin EV. The COG database: a tool forgenome-scale analysis of protein functions and evolution. Nucleic Acids Res 2000;28:33–6.

[23] Thompson JD, Higgins DG, Gibson TJ. CLUSTAL W: improving the sensitivity ofprogressive multiple sequence alignment through sequence weighting, position-

specific gap penalties and weight matrix choice. Nucleic Acids Res 1994;22:4673–80.

[24] Retief JD. Phylogenetic analysis using PHYLIP. Methods Mol Biol 2000;132:243–58.

[25] Choi JH, Jung HY, Kim HS, Cho HG. PhyloDraw: a phylogenetic tree drawingsystem. Bioinformatics 2000;16:1056–8.

[26] Zheng D, Feeney GP, Kille P, Hogstrand C. Regulation of ZIP and ZnT zinctransporters in zebrafish gill: zinc repression of ZIP10 transcription by an intronicMRE cluster. Physiol Genomics 2008;34:205–14.

[27] Saydam N, Steiner F, Georgiev O, Schaffner W. Heat and heavy metal stresssynergize to mediate transcriptional hyperactivation by metal-responsivetranscription factor MTF-1. J Biol Chem 2003;278:31879–83.

[28] Sun L, Yu Y, Huang T, An P, Yu D, Yu Z, et al. Associations between ionomic profileand metabolic abnormalities in human population. PLoS One 2012;7:e38845.

[29] Zhao T, Chen T, Qiu Y, Zou X, Li X, Su M, et al. Trace element profiling usinginductively coupled plasma mass spectrometry and its application in anosteoarthritis study. Anal Chem 2009;81:3683–92.

[30] Bittel D, Dalton T, Samson SL, Gedamu L, Andrews GK. The DNA binding activityof metal response element-binding transcription factor-1 is activated in vivoand in vitro by zinc, but not by other transition metals. J Biol Chem 1998;273:7127–33.

[31] Pinilla-Tenas JJ, Sparkman BK, Shawki A, Illing AC, Mitchell CJ, Zhao N, et al. Zip14is a complex broad-scope metal-ion transporter whose functional propertiessupport roles in the cellular uptake of zinc and nontransferrin-bound iron. Am JPhysiol Cell Physiol 2011;301:C862–71.

[32] Wang K, Zhou B, Kuo YM, Zemansky J, Gitschier J. A novel member of a zinctransporter family is defective in acrodermatitis enteropathica. Am J Hum Genet2002;71:66–73.

[33] Wang F, Kim BE, Dufner-Beattie J, Petris MJ, Andrews G, Eide DJ. Acrodermatitisenteropathica mutations affect transport activity, localization and zinc-responsive trafficking of the mouse ZIP4 zinc transporter. Hum Mol Genet2004;13:563–71.

[34] Ryu MS, Lichten LA, Liuzzi JP, Cousins RJ. Zinc transporters ZnT1 (Slc30a1), Zip8(Slc39a8), and Zip10 (Slc39a10) in mouse red blood cells are differentiallyregulated during erythroid development and by dietary zinc deficiency. J Nutr2008;138:2076–83.

[35] Wang B, Schneider SN, Dragin N, Girijashanker K, Dalton TP, He L, et al. Enhancedcadmium-induced testicular necrosis and renal proximal tubule damage causedby gene-dose increase in a Slc39a8-transgenic mouse line. Am J Physiol CellPhysiol 2007;292:C1523–35.

[36] Taylor KM, Morgan HE, Smart K, Zahari NM, Pumford S, Ellis IO, et al. Theemerging role of the LIV-1 subfamily of zinc transporters in breast cancer. MolMed 2007;13:396–406.

[37] Dalton TP, He L, Wang B, Miller ML, Jin L, Stringer KF, et al. Identification of mouseSLC39A8 as the transporter responsible for cadmium-induced toxicity in thetestis. Proc Natl Acad Sci USA 2005;102:3401–6.

[38] He L, Girijashanker K, Dalton TP, Reed J, Li H, Soleimani M, et al. ZIP8, member ofthe solute-carrier-39 (SLC39) metal-transporter family: characterization oftransporter properties. Mol Pharmacol 2006;70:171–80.

[39] Kelleher SL, Velasquez V, Croxford TP, McCormick NH, Lopez V, MacDavid J.Mapping the zinc-transporting system in mammary cells: molecular analysisreveals a phenotype-dependent zinc-transporting network during lactation. J CellPhysiol 2012;227:1761–70.

[40] Aydemir TB, Sitren HS, Cousins RJ. The zinc transporter Zip14 influences c-Metphosphorylation and hepatocyte proliferation during liver regeneration in mice.Gastroenterology 2012.

[41] Singh RP, Kumar S, Nada R, Prasad R. Evaluation of copper toxicity in isolatedhuman peripheral blood mononuclear cells and it's attenuation by zinc: ex vivo.Mol Cell Biochem 2006;282:13–21.

[42] Gaither LA, Eide DJ. The human ZIP1 transporter mediates zinc uptake in humanK562 erythroleukemia cells. J Biol Chem 2001;276:22258–64.

[43] Gaither LA, Eide DJ. Functional expression of the human hZIP2 zinc transporter.J Biol Chem 2000;275:5560–4.

[44] Wang F, Dufner-Beattie J, Kim BE, Petris MJ, Andrews G, Eide DJ. Zinc-stimulatedendocytosis controls activity of the mouse ZIP1 and ZIP3 zinc uptake transporters.J Biol Chem 2004;279:24631–9.

[45] Liuzzi JP, Bobo JA, Lichten LA, Samuelson DA, Cousins RJ. Responsive transportergenes within the murine intestinal-pancreatic axis form a basis of zinchomeostasis. Proc Natl Acad Sci USA 2004;101:14355–60.

[46] Weaver BP, Dufner-Beattie J, Kambe T, Andrews GK. Novel zinc-responsive post-transcriptional mechanisms reciprocally regulate expression of the mouseSlc39a4 and Slc39a5 zinc transporters (Zip4 and Zip5). Biol Chem 2007;388:1301–12.

[47] Guo CH, Chen PC, Yeh MS, Hsiung DY, Wang CL. Cu/Zn ratios are associated withnutritional status, oxidative stress, inflammation, and immune abnormalities inpatients on peritoneal dialysis. Clin Biochem 2011;44:275–80.

[48] Cragg RA, Phillips SR, Piper JM, Varma JS, Campbell FC, Mathers JC, et al.Homeostatic regulation of zinc transporters in the human small intestine bydietary zinc supplementation. Gut 2005;54:469–78.

[49] Dufner-Beattie J, Langmade SJ, Wang F, Eide D, Andrews GK. Structure, function,and regulation of a subfamily of mouse zinc transporter genes. J Biol Chem 2003;278:50142–50.

[50] McMahon RJ, Cousins RJ. Regulation of the zinc transporter ZnT-1 by dietary zinc.Proc Natl Acad Sci USA 1998;95:4841–6.