Biochimica et Biophysica Acta 1823 (2012) 1416–1425

Contents lists available at SciVerse ScienceDirect

Biochimica et Biophysica Acta

j ourna l homepage: www.e lsev ie r .com/ locate /bbamcr

The taste of heavy metals: Gene regulation by MTF-1☆

Viola Günther 1, Uschi Lindert 1, Walter Schaffner ⁎Institute of Molecular Life Sciences, University of Zürich, Winterthurerstrasse 190, CH-8057 Zürich, Switzerland

☆ This article is part of a Special Issue entitled: Cell B⁎ Corresponding author. Tel.: +41 44 635 3150; fax:

E-mail address: walter.schaffner@imls.uzh.ch (W. Sc1 Equal contribution.

0167-4889/$ – see front matter © 2012 Elsevier B.V. Aldoi:10.1016/j.bbamcr.2012.01.005

a b s t r a c t

a r t i c l e i n f o

Article history:Received 16 November 2011Received in revised form 8 January 2012Accepted 11 January 2012Available online 20 January 2012

Keywords:MTF-1Heavy metal homeostasisZincCopperMetallothioneinTranscription

The metal-responsive transcription factor-1 (MTF-1, also termed MRE-binding transcription factor-1 or metalregulatory transcription factor-1) is a pluripotent transcriptional regulator involved in cellular adaptation tovarious stress conditions, primarily exposure to heavy metals but also to hypoxia or oxidative stress. MTF-1 isevolutionarily conserved from insects to humans and is the main activator of metallothionein genes, whichencode small cysteine-rich proteins that can scavenge toxic heavy metals and free radicals. MTF-1 hasbeen suggested to act as an intracellular metal sensor but evidence for direct metal sensing was scarce.Here we review recent advances in our understanding of MTF-1 regulation with a focus on the mechanismunderlying heavy metal responsiveness and transcriptional activation mediated by mammalian or DrosophilaMTF-1. This article is part of a Special Issue entitled: Cell Biology of Metals.

© 2012 Elsevier B.V. All rights reserved.

1. Introduction

The term heavy metal comprises a number of essential and non-essential metals. Among the latter cadmium, mercury and lead aretoxic even in trace amounts. Although zinc and copper, which are essen-tial heavy metals, are integral parts of proteins, notably enzymes andtranscription factors, an excess of these metals is also toxic. Therefore,elaborate systems to import, sequester, store, transport and expelmetals have evolved. Important players in metal homeostasis are themetallothioneins. These small proteins with a high content of cysteinescan bind and thereby sequester heavy metals. Metallothioneins werediscovered in 1957 by Margoshes and Vallee as cadmium-binding pro-teins in preparations of horse kidney [1]. Transcription of metallothio-nein genes is upregulated in response to different stimuli, especiallyheavy metals, but also oxidative stress or stress hormones (glucocorti-coids). A hallmark of the promoters/enhancers of most metallothioneingenes are short DNA sequence motifs termed metal response elements(MREs); many also harbor antioxidant response elements (AREs) andglucocorticoid response elements (GREs) [2–4]. MREs were originallydiscovered as conspicuous DNA sequence motifs present in multiplecopies in the promoter region of the mouse metallothionein 1 gene. Intransfection experiments, a reporter gene driven by a synthetic, MRE-containing promoter became responsive to zinc and cadmium [5,6].MREs, often in multiple copies, are typically present in the upstreamregulatory sequences of other heavy metal-responsive genes. The

iology of Metals.+41 44 635 6830.haffner).

l rights reserved.

identification of these MREs, that share the core consensus sequenceTGCRCNC (R=A or G, N=any nucleotide), suggested the existence ofa specific transcription factor that regulates metallothionein expressionin response to metals [7]. Indeed in 1988 a MRE-binding protein wasidentified by electrophoretic mobility shift and methylation interfer-ence studies [8,9]. It is bound to its cognate DNA motif in a zinc depen-dentmanner [8] andwas termedMTF-1 (forMRE-binding transcriptionfactor-1, more recently also metal-responsive transcription factor-1 ormetal regulatory transcription factor-1). In 1993 the cDNA of mouseMTF-1 was cloned which revealed it as a zinc finger protein [10].HumanMTF-1 with a length of 753 amino acids was cloned soon there-after and found to be highly similar but slightly longer at the C-terminusthan mouse MTF-1 [11,12]. A secondMRE-binding protein unrelated toMTF-1, termed MTF2 or ZiRF1, was found in a yeast one-hybrid assay[13]. Its significance, if any, for metal homeostasis remains unclear. Sev-eral earlier reviews have addressed aspects of MTF-1 function [14–18].Herewe attempt a presentation of the current state of knowledge incor-porating new data that has arisen in the meantime.

MTF-1 is evolutionarily conserved from insects to mammals, it sofar has been characterized in human, mouse, capybara, fish and Dro-sophila [10,11,19–24], where it is typically present as a single copygene. Database searches also revealed orthologs of MTF-1 in the ge-nomes of other vertebrates and insects. Genes with an array of zincfingers typical for MTF-1 can even be tracked down in invertebratessuch as sea urchin, sea anemone and hydra (K. Steiner and W.S.),but their role in metal homeostasis has not been addressed experi-mentally so far. While metallothionein genes are also present in C.elegans and in yeast, no MTF-1 ortholog has been identified in theseorganisms. In yeast metal-responsive transcription factors unrelatedto MTF-1 were identified [25]. In C. elegans the GATA-type

1417V. Günther et al. / Biochimica et Biophysica Acta 1823 (2012) 1416–1425

transcription factor ELT-2 is essential for metallothionein transcrip-tion, but data about metal-sensing regulators are still missing [26].

MTF-1 not only plays a central role in the homeostasis and detoxi-fication of heavymetals, but also helps to protect the organism againsthypoxia and oxidative stress [27–32]. Furthermore, in mammals MTF-1 is essential for embryonic liver development, andmice lacking MTF-1 die at embryonic stage due to liver degeneration [33,34]. In contrastto this, Drosophila lacking MTF-1 are viable under normal conditionsbut are markedly sensitive to heavy metal stress [35].

Considerable efforts have been undertaken to understand how themetal-dependent activity of MTF-1 is regulated. In a subset of studies,transcript and protein levels of MTF-1 were reported to increase uponmetal treatment [36–38]. However, MTF-1 under the control of a con-stitutive promoter is still fully functional, suggesting that its functionis mainly regulated post-translationally [39]. Indeed MTF-1's activitycan be regulated at several steps: nuclear–cytoplasmic shuttling,DNA-binding, and the interaction with other transcriptional coactiva-tors. Under normal conditions mammalian MTF-1 localizes both tothe nucleus and the cytoplasm but accumulates in the nucleus upondiverse stresses (Fig. 1), including heavy metal overload, heat shockand oxidative stress [40,41]. Zinc sensing by MTF-1 can be mediatedby the zinc fingers and a metal-responsive activation domain (see“Functional domains of MTF-1”), whereas response to other stressesis most likely indirect, via zinc released from metallothioneins andother proteins (Fig. 1). After DNA-binding MTF-1 recruits differentco-activators and often relies on other transcription factors for coor-dinated target gene expression (see “Interaction partners of MTF-1”and Table 2).

p300

Sp1

MRE

cytosol

nucleus

phosphorylationdephosphorylation

P

P

MTF-1

MTF-1

MTF-1

Crm1

MTF-1

Fig. 1. Overview of mammalian MTF-1 regulation. Under normal conditions MTF-1 is shuttliaction of MTF-1 with Crm1 [52], which mediates nuclear export of proteins that carry a leurelease of zinc from metallothioneins upon cadmium load or oxidative stress (H2O2), as suggmodulated by phosphorylation. Upon zinc binding MTF-1 shuttles to the nucleus where it bsus sequence TGCRCNC (R=A or G; N=any nucleotide). MTF-1 interacts with other transc

2. Functional domains of MTF-1: an unexpected human superiority

Fig. 2A shows an overview of the domain composition of humanMTF-1. Its DNA-binding domain is composed of six zinc fingers ofthe Cys2His2-type. Since MTF-1 shows an increased DNA-bindingupon zinc supplementation [8,42] and individual zinc fingers havedifferent zinc binding affinities [43,44], this domain was suggestedto mediate the intrinsic zinc sensing of MTF-1. Zinc fingers 5 and 6can contribute to specificity but are dispensable for MRE-bindingper se, as suggested by alkylation of individual zinc fingers withNEM, proteolysis and deletion experiments [17,45,46]. Domain ex-change experiments between MTF-1 and another zinc finger tran-scription factor (Sp1) suggested a role for zinc finger 1 of MTF-1 inzinc-dependent DNA binding [47]. Since no crystal structure of theprotein/DNA complex is available, interactions between the zinc fin-ger residues and MRE bases are extrapolated from other zinc finger/DNA complexes [17].

Some of the “linker” sequences between the zinc fingers of MTF-1do not conform to the linker consensus sequence TGE[K/R]P which isfound in approximately 40% of all multi-Cys2His2 zinc finger domains[48]. Mutation of the non-conventional linker between zinc fingers 1and 2 of mouse MTF-1 led to permanent nuclear localization, DNA-binding and constitutive transcriptional activity [49]; the same muta-tion in human MTF-1 caused constitutive nuclear localization, butDNA binding and transcriptional activity remained zinc-responsive(UL and WS, unpublished). This and other results (discussed below)indicate that the mechanism of metal inducibility of human MTF-1is more elaborate and/or more robust than the one of the mouse.

MT

Cd2+

Zn2+

MT

H2O2

transcription of target genes

TBP

coactivator/Mediator

general transcr.machinery

ng between the cytoplasm and the nucleus, with export most likely mediated by inter-cine-rich nuclear export signal. MTF-1 can be activated directly by zinc or indirectly byested by cell free transcription experiments [123]. In addition, activity of MTF-1 can beinds to its cognate DNA motif, termed MRE (metal-response element) with the consen-ription factors (e.g. Sp1) and coactivators (e.g. p300) to drive target gene expression.

1

zinc fingers139 316

632 638

CQCQCACcysteinecluster

335 346

SLCLSDLSLLSTSLYLSELGLLST

NESnuclear export

133 139

KRKEVKR aNLS

nuclear import

acidic pro-rich ser/thr-rich330 620

753

160 228

NLSnuclearimport

DNA bindingzinc sensing

homodimerization

transactivationactivation domains

zinc sensingp300 binding

hm

285

791

copper sensingcysteine cluster

CNCTNCKCDQTKSCHGGDC

547 565

613 630

MNQNIEDVALLLQNLASMNES1

nuclear export

763 777

VVICIKTLQALRKVLNES2

nuclear export

112

DNA bindingzinc fingers

1

transactivationactivation domain

112 202

NLSnuclear import

292 300

KGRKKRPLKNLS

nuclear import

540352

metal sensingautoinhibition

MT-like

A

B

C-terminus in MTF-1 of:capybara mouse, rat

Fig. 2. Domain structure of human vs. Drosophila MTF-1. A. Functional domains of human MTF-1. Depicted are the six Cys2His2-type zinc fingers, three transcriptional activationdomains, the nuclear localization signal spanning zinc fingers 1–3 (NLS), the auxiliary nuclear localization signal (aNLS), the nuclear export signal (NES) overlapping the acidic ac-tivation domain and the C-terminal cysteine cluster. The protein sequences of important domains, including the deviant mouse NES sequence, are shown. Cysteine residues areindicated by purple dashes. The C-terminus of mouse and capybara MTF-1 is indicated by arrows. B. Domain organization of DrosophilaMTF-1. Schematic overview of the functionaldomains of Drosophila MTF-1, including the six zinc fingers, the activation domain, nuclear localization signals (NLS) and nuclear export signals (NES). The metallothionein-like(MT-like) domain is characterized by a high content of cysteines; cysteine residues are indicated by purple dashes.

1418 V. Günther et al. / Biochimica et Biophysica Acta 1823 (2012) 1416–1425

Transcriptional activation domains were functionally classifiedaccording to their amino acid composition [50]. By fusing segmentsof mouse MTF-1 to the Gal4 DNA-binding domain, three transcrip-tional activation domains were identified in the C-terminal half: anacidic, a proline-rich and a serine/threonine-rich domain [51]. Theacidic activation domain is the strongest and harbors a major deter-minant of metal inducibility: when fused to a heterologous DNA-binding domain this chimeric transcription factor shows a relativelystrong inducibility by zinc, though still not quite as high as wildtype human MTF-1. However the response via this domain was onlyseen in three of six cell lines investigated and thus is not as robustas that of complete human MTF-1, which is metal responsive in allmammalian cell types tested so far [52].

The subcellular distribution of human MTF-1 is regulated by dis-tinct motifs: a non-conventional nuclear localization signal (NLS)and a nuclear export signal (NES) [40,52]. The former overlaps withzinc fingers 1–3, the latter is embedded in the acidic activation do-main and resembles classical, leucine-rich NESs that mediate the in-teraction with the export receptor Crm1. In accordance with thisnotion, treatment of cells with leptomycin B (LMB), an established in-hibitor of Crm1, leads to nuclear accumulation of MTF-1 [52]. Thesefindings support the concept of MTF-1 as a nuclear-cytoplasmic shut-tling protein with balanced import and export in non-stressed cells.The nuclear-cytoplasmic distribution of mouse MTF-1 was found tobe coupled to activity (see above). By contrast, the significance ofthe NES for human MTF-1 function is not clear. Inhibition of exportby LMB-treatment only slightly affected zinc-induced reporter tran-script levels, again this points to a more robust metal responsivenessof human MTF-1 compared to mouse MTF-1 [52]. The NES sequenceis not only required for export from the nucleus but also is an es-sential part of the acidic activation domain in human MTF-1. Thissuggests a model in which the acidic activation domain providesmutually exclusive binding to either the export receptor or a tran-scriptional cofactor.

In reporter assays, human MTF-1 exhibits a much stronger metalresponse than mouse MTF-1, irrespective of whether tested inhuman or mouse cells [53]. The low metal inducibility of mouseMTF-1 seems to be characteristic for rodents, since MTF-1 from capy-bara, the largest living rodent, behaves like mouse MTF-1 [24]. A se-quence comparison revealed that the extended form of MTF-1 is theprimordial form that is typical for all mammals except rodents. Theproteins of mouse and capybara became truncated at the C-terminusby premature terminationmutations (Fig. 2A). However, unexpected-ly, the critical determinant for the different metal inducibility of ro-dent and human MTF-1 turned out to be elsewhere: in contrast to itshuman counterpart (see above) the isolated acidic activation domainof mouse MTF-1 does not show metal responsiveness. However,mouse MTF-1 can readily be “humanized” by changing three aminoacids in the NES/acidic activation domain in mouse MTF-1. Such anMTF-1, despite being still truncated by 78 amino acids at the C-terminus, displays the high metal inducibility characteristic forhuman MTF-1. Conversely, “mousifying” the human NES decreasesthe metal response, showing that the NES sequence is not only re-quired for export from the nucleus but also is an essential part of theacidic activation domain in human MTF-1 [52]. The low metal re-sponse of rodent MTF-1 relative to human MTF-1 is also observed intransfected rodent cells and on rodent metallothionein reporter con-structs, i.e., it is independent of the host cell or reporter system used,and we have speculated that heavy metal stress may have been lessof an issue during rodent evolution [24]. While we cannot excludethat an unknown factor(s) contributes to metal homeostasis in ro-dents, it is clear that themetal response of humanMTF-1 is intrinsical-ly more robust. The fact that humans have at least 10 metallothioneingenes (plus a few pseudogenes) versus 4 in the mouse is consistentwith the idea that long-lived organisms are particularly vulnerableto heavy metal accumulation and -toxicity.

The C-terminal segment of vertebrate MTF-1 harbors a conservedcysteine-rich motif (CQCQCAC) that was shown to be necessary for

′

A

B

0

1

2

bit

s

5′

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 3′

0

1

2

bit

s

5′

1 2 3

A

4 5 6 7 8 9 10 11 12 13 14 15 3

Fig. 3. Consensus sequence of human and Drosophilametal response elements. Analysisof MREs located in the promoter region of target genes revealed that mammalian (A)and Drosophila (B) MTF-1 share the same core consensus and only seem to differ insome bases flanking the MREs. The consensus sequences were determined by align-ment of 41 MREs originating from 16mammalian target genes and 36 MREs originatingfrom 9 target genes in Drosophila using WebLogo [125]. Note that thymine residuespreceding the MRE core consensus occur in a subset of Drosophila MREs but are veryrare in mammalian MREs. Nevertheless, such MREs were selected by human MTF-1in an in vitro selection assay starting with random oligonucleotides under low zinc con-ditions [126], and might therefore confer particularly strong MTF-1 binding. Mamma-lian MREs are generally in a GC-rich background.

1419V. Günther et al. / Biochimica et Biophysica Acta 1823 (2012) 1416–1425

metal-induced transcription [54,55]. Studies in our lab revealed thatthis cysteine cluster mediates homodimerization of MTF-1. Most like-ly only one partner of dimeric MTF-1 binds to DNA, as no conservedspacing between directly or divergently orientated MREs is presentin the promoter and enhancer regions of different MTF-1 targetgenes (K. Steiner and W.S.). The dimerized molecule might offer aplatform for the recruitment of transcriptional cofactors (Fig. 1). Al-though it was tempting to speculate that the cysteines are involvedin metal-sensing, i.e., that homodimerization is regulated by metalbinding, our data showed that dimerization is constitutive [56].Taken together, the most recent studies bring the acidic activation do-main into focus as a major mediator of metal responsiveness ofhuman MTF-1, with homodimerization being a prerequisite.

Recently a sumoylation site at position Lys627 and a SUMO-interacting motif 621VPVIII626 was identified for mouse MTF-1.These are conserved among vertebrates. The sumoylated form ofMTF-1 resides in the cytoplasm, where it probably forms SUMO-dependent complexes [57].

Drosophila also contain a MTF-1 homolog. Besides the 791 aminoacid form which was cloned and characterized [21] gene annotationprograms also predict an alternative Drosophila MTF-1 splice formencoding a 1006 amino acid protein, which is conserved among dro-sophilids. In addition, on the complementary strand an open readingframe that encodes a MADF domain containing protein is present.However, both the 1006 amino acid variant of MTF-1 and the openreading frame for the putative MADF DNA-binding protein cannotbe essential: in a genetic background that precludes their expressionand solely relies on a cDNA encoding the standard 791 amino acidform of MTF-1, all investigated phenotypes were completelyrescued.

Drosophila MTF-1 is composed of domains with similar functionsas those found in humanMTF-1, but their primary sequence and loca-tion within the protein sometimes differ (Fig. 2B). The zinc finger re-gion is the only domain with clear sequence homology to humanMTF-1 (81% similarity, 68% identity). A nuclear localization functionis harbored within zinc fingers 1–3 [58]. Mammalian and DrosophilaMTF-1 bind to the same core MRE consensus (Fig. 3). There is littleif any protein sequence similarity outside of the zinc fingers, but nev-ertheless mammalian and Drosophila MTF-1 can largely cross-complement each other when tested in the respective knock-outbackground [16].

In contrast to human MTF-1, Drosophila MTF-1 contains an activa-tion domain that is not metal-responsive but shows strong constitu-tive activity when fused to a heterologous DNA-binding domain.Two domains of Drosophila MTF-1 seem to be involved in metal sens-ing: a cysteine cluster “CNCTNCKCDQTKSCHGGDC” [59] and a C-terminal cysteine-rich “metallothionein-like” domain [58]. As for itsmammalian ortholog (see below) regulation of Drosophila MTF-1 ac-tivity is mediated in part via phosphorylation. Four phosphorylationsites, located in the zinc finger region, were identified in a phospho-peptide screen carried out with Drosophila cells [60]. Mutation ofthese sites strongly affected the expression of metallothionein B butnot of metallothionein A, indicating a target gene discriminationeven between members of the metallothionein gene family [58], dis-cussed also below [85].

Furthermore, the C-terminal domain plays an essential role inthe regulation of Drosophila MTF-1. First, it autorepresses MTF-1's transcriptional activity, likely via inhibiting DNA-binding, andsecond it mediates degradation of MTF-1 after prolonged heavy-metal exposure, presumably to counteract over activation of tar-get genes [58]. The importance of the C-terminus in the regulationof MTF-1 becomes obvious when C-terminal truncations of MTF-1are introduced into the Drosophila genome: such transgenic fliessuffer from female sterility, a markedly reduced lifespan and adistorted wing development, most likely as a result of excessiveexpression of target genes [58].

3. MTF-1 target genes: too many responses for one factor?

The search for MTF-1 target genes other than metallothioneingenes has been performed in human, mouse and Drosophila systems(Table 1). Mammalian target genes with a role in metal homeostasisthat are induced upon zinc load include Slc30a1 and Slc30a2 (solutecarrier family 30 member 1 and 2, respectively). These encode thezinc exporters ZnT-1 and ZnT-2, respectively [61,62]. While MREsare typically located upstream, MTF-1 dependent expression of ZnT-2 is mediated by an MRE located 53 bp downstream of the transcrip-tion start [61]. Additionally MTF-1 represses transcription of Slc39a10[63], which was subsequently shown to encode the zinc importerZip10 [64]. In this case binding of MTF-1 to an MRE located 17 bpdownstream of the transcription start results in transcriptional re-pression via RNA Pol II stalling [65].

Even though MTF-1 is not inducible by iron, a connection to thisessential trace metal is given by the fact that expression of ferroportin1, an iron export protein which shows also modest zinc export activ-ity, is induced directly via MTF-1 [66]. Curiously, expression of hepci-din which is an inhibitor of ferroportin 1 was also shown to beinduced by MTF-1 in a zinc-dependent manner via four MREs in itspromoter [67]. Taken at face value, these data seem contradictory,but the induction thresholds or kinetics may differ.

A role for MTF-1 (and Sp1) was reported in copper-dependentupregulation of the gene encoding the prion protein PrP [68], whichhad long been implicated in copper homeostasis [69–72]. MTF-1also activates expression of β-synuclein [73], another protein relevantto neurodegeneration and was postulated to counteract the aggrega-tion of α-synuclein [74,75], which is typically found in Parkinson'sdisease.

The role of MTF-1 in the defense against oxidative stress is under-lined by the MTF-1 dependent expression of the selenoprotein 1 gene(Sepw1) which encodes a glutathione-binding protein with antioxi-dant activity [63]. However, MTF-1 is also involved in transcriptional

Table 1Major target genes of mammalian and Drosophila MTF-1.

Gene/protein Organism Regulation via MTF-1 GO annotation: “biological process” (selected terms) Ref.

MT/metallothionein Mouse/human/Drosophila

Basal and metal induced Cellular copper homeostasis, cellular response to zinc,negative regulation of growth, cellular response to cadmium ion.

[10][21][42]

Slc30a1/zinc transporter-1 Mouse Basal and Zn/Cd induced Zinc ion transport, cellular zinc ion homeostasis, calcium ion import,detoxification of cadmium ion, in utero embryonic development,

[63]

Slc30a2/zinc transporter-2 Mouse Zn induced Zinc ion transport [61]Slc39a10/zinc transporter Zip10 Mouse Basal and Zn/Cd induced

repressionZinc ion transport [63]

[65]Slc40a1/ferroportin-1 Mouse Zn/Cd induced Iron ion transport, endothelium development [66]HAMP/hepcidin Human Zn/Cd induced Cellular iron homeostasis, immune response, defense response to

bacteria and fungi[67]

PRNP/major prion protein Human Cu induced Cellular copper ion homeostasis, response to oxidative stress, axonguidance, negative regulation of activated T cell proliferation

[68]

SNCB/β-synuclein Human Cu induced Dopamine metabolic processing, negative regulation of neuronapoptosis

[73]

Sepw1/selenoprotein W Mouse Basal Cellular redox homeostasis [63]SELH/selenoprotein H Mouse/human Zn induced repression Cellular redox homeostasis [76]Txnrd2/thioredoxin reductase 2 Mouse Zn induced repression Cellular redox homeostasis, response to oxygen radical [76]GCLC/glutamate–cysteine ligase,catalytic subunit

Mouse Basal Glutathione biosynthetic process, cellular redox homeostasis,response to heat

[33]

PlGF/placenta growth factor Mouse/human Hypoxia induced Angiogenesis, cell–cell signaling, cell differentiation, positiveregulation of cell proliferation

[27][78]

Ndrg1/N-myc downstream-regulatedgene 1

Mouse Cd induced Cellular response to hypoxia, mast cell activation, peripheralnervous system myelin maintenance

[63]

DDAH1/N(G),N(G)-dimethylargininedimethylaminohydrolase 1

Human Basal Nitric oxide mediated signal transduction, positive regulationof angiogenesis

[80]

AFP/α-fetoprotein Mouse Basal Liver development, response to organic substance, sexualreproduction, pancreas development,

[77]

C/EBPα/CCAAT/enhancer-bindingprotein-α

Mouse Basal and Zn induced Liver development, regulation of transcription DNA-dependent,cell differentiation, regulation of cell proliferation, cellular responseto lithium ion

[77]

LCN1/tear lipocalin Mouse Basal Response to stimulus, sensory perception of taste [77]Csrp1/cysteine and glycine-richprotein 1

Mouse Cd induced Actin cytoskeleton organization [63]

ZnT35C Drosophila Basal and Zn/Cd induced Response to zinc ion, transmembrane transport [82]CG10505 Drosophila Zn/Cd/Cu induced Transmembrane transport, response to copper, response

to cadmium[82]

Fer2LCH, Fer1HCH/ferritin Drosophila Zn/Cd/Cu induced Cellular iron ion homeostasis, oxidation-reduction process,bristle development

[82]

CG1886/ATP7 Drosophila Cu induced Copper ion export, larval development, developmental pigmentation,ATP biosynthetic process

[86]

CG7459/Ctr1B Drosophila Basal and low Cu induced Copper ion transmembrane transport, response to metal ion [88]

1420 V. Günther et al. / Biochimica et Biophysica Acta 1823 (2012) 1416–1425

repression of the redox sensing selenoprotein H and the thioredoxinreductase 2 via MREs located downstream of the transcription start[76]. The gene for the catalytic subunit of glutamate–cysteine ligase(GCLC), an essential enzyme in glutathione biosynthesis, containsMREs in its promoter region and is downregulated in MTF-1 null mu-tant mouse embryos [33]. However the reduced transcript levelsmight primarily be a result of liver decay as glutathione levels remainhigh in embryos at earlier stage [77].

A link to the hypoxic stress response was established by the find-ing that MTF-1 contributes to the expression of placental growth fac-tor (PlGF) [27], a member of the vascular endothelial growth factorfamily. MTF-1 activates the PlGF promoter in concert with the tran-scription factor NF-κB upon hypoxia [78]. MTF-1 also drives thecadmium-induced expression of N-myc downstream regulated gene1 (Ndrg1) [63], which encodes a protein of unknown molecular func-tion. Interestingly however Ndrg1 is induced by several stress condi-tions, is overexpressed in many types of cancer and its expressionunder hypoxic conditions is dependent on hypoxia-inducible factor1 alpha (HIF-1α) [79]. A polymorphism in the promoter of the geneencoding dimethylarginine dimethylaminohydrolase 1 (DDAH1),that disrupts an MRE and leads to loss of MTF-1 binding, results in de-creased DDAH1 mRNA levels and is associated with an increased riskof thrombosis stroke and coronary heart disease [80].

Target genes that might explain the lethal liver degeneration ob-served in MTF-1 null mutant mice are α-fetoprotein (AFP) and C/EBPα [77], both implicated in liver development in separate studies.

Other genes that were shown to be regulated via MTF-1 withoutan obvious link to a stress response include the gene for cysteine-and glycine-rich protein 1 (Csrp1) [63], Krüppel-like factor 4 (KLF4),hepatitis A virus cellular receptor 1 (HAVCR1), complement factor B(CFB) [81] and tear lipocalin (LCN1) [77].

In insects, the search for MTF-1 target genes in the vinegar fly D.melanogaster yielded both similar and different targets: again thefamily of metallothionein genes emerged as major targets, as didZnT35C a zinc efflux transporter, CG10505 a homolog of the “yeastcadmium factor” and ferritin genes [35,82–84]. The major role of fer-ritins is in iron homeostasis, but they were also implicated in the se-questration of other metals including, Zn, Cu, Cd [85]. MTF-1 alsoincreases the expression of the copper exporter DmATP7 in larvalmidgut in response to copper load [86]. DmATP7 is the Drosophilaortholog of the human copper transporters ATP7A and ATP7B. Muta-tions in these result in Menkes and Wilson's disease, respectively, asa result of severe distortions of copper homeostasis [87]. AlthoughMTF-1 typically induces target genes in response to metal load, inDrosophila it also induces transcription of the intestinal copper im-porter Ctr1B in response to copper starvation and is also needed forthe repression of Ctr1B upon copper load [88]. Of note, in contrastto mammalian copper importers which are regulated post-transcriptionally, activity of Drosophila Ctr1B is mostly regulated atthe transcriptional level [89].

DrosophilaMTF-1, which also influences the expression of the cop-per exporter DmATP7 [86], can be considered to be the main regulator

1421V. Günther et al. / Biochimica et Biophysica Acta 1823 (2012) 1416–1425

of copper homeostasis in the fly. The role of MTF-1 in copper homeo-stasis in mammals is not well established. Copper availability doesnot regulate the expression of the mammalian copper transportersCtr1, ATP7A and ATP7B, rather their intracellular localization is chan-ged upon altered copper levels [90]. In line with mammalian MTF-1having a minor role in copper homeostasis, zinc and cadmium inducea very fast and robust transcription of MRE-containing reporter genesin all cell types tested, while a response to copper is restricted to spe-cific cell types (own observations). In mice a number of cadmium-responsive genes were also found to be regulated independently ofMTF-1 and several of these are involved in glutathione metabolism.MTF-1 null mutant cells are extremely sensitive to cadmium whenglutathione synthesis is inhibited by treatment with buthionine sul-foximide. This is strong evidence for two largely separate branchesof cadmium detoxification, namely, one via MTF-1 and the other viaglutathione [63]. Post-transcriptional regulation of a group of zinctransporters via a microRNA cluster was recently shown to representanother MTF-1 independent mechanism in heavy metal homeostasis[91].

It seems paradoxical that baker's yeast, for example, has three dif-ferent heavy metal-responsive transcription factors (Mac1, Ace1,Zap1) whereas higher organisms seem to rely only on MTF-1. Howev-er, specificity of MTF-1 can be determined by the promoter architec-ture of target genes and the interaction of different cofactors withspecific domains of MTF-1, as suggested by mutagenesis experiments[58,59,88,92,93]. Furthermore, posttranscriptional regulation of metaltransporters also contributes to metal homeostasis.

4. Interaction partners of MTF-1: for better or worse

Considerable efforts have gone into the identification of MTF-1'sinteraction partners (Table 2). Identified partners include basaltranscriptional coactivators such as TFIID and Mediator, as shown inD. melanogaster [92]. This study also revealed an interesting aspect:the promoters of Drosophila metallothioneins A and B, which showpreferential response to copper and cadmium, respectively [84], alsoshow differences in MTF-1 dependent coactivator recruitment [92].

An interactome study in Drosophila uncovered two novel interac-tion partners [94]: Dpy-30L1 and Dpy-30L2. They are named this be-cause of their sequence similarity to Dumpy-30 from C. elegans, asmall protein involved in chromatin modification. Dpy-30L1 inter-feres with the binding of Drosophila MTF-1 to MREs. This interactionis apparently insect-specific since it was not observed for human ormouse MTF-1 [95].

Mouse MTF-1 was found to form a zinc-induced complex with thetranscription factor Sp1 and the coactivator/histone acetyltransferasep300. This interaction was mediated by the acidic activation domain

Table 2Main interaction partners of mammalian and Drosophila MTF-1.

Protein Organism Comments

TFIID complex Drosophila Subunits mostly inhibit initiation of transcriMediator complex Drosophila Different subunits are recruited to the promDpy30L1 (and Dpy30L2) Drosophila Direct interaction; negative effect by inhibit

Sp1 Mouse/human Direct interaction; complex results: activate

p300/CBP Mouse/human Direct interaction; contributes to inductionPTEN Human/mouse Direct interaction in the cytoplasm; activateHIF-1α Mouse Direct interaction; cooperates with MTF-1 tNrf1 Mouse Nrf1 and MTF-1 cooperate to activate basalHSF1 Human MTF-1 reduces metal-induced HSF1-depend

USF1 Mouse MTF-1 and USF1 cooperate to activate basalNF1 Mouse Contradicting results: NF1 is reported to eit

C/EBPα Human Direct interaction; C/EBPα stimulates MT-2a

[*] Own observations.

in mouse MTF-1 [93] and recruitment of p300 lead to chromatinremodeling at the mouse metallothionein 1 promoter [96]. Interest-ingly, chromium(VI) inhibits the formation of the p300/Sp1/MTF-1complex on the mouse metallothionein 1 promoter and thereby im-pairs RNA polymerase recruitment, suggesting a mechanism of chro-mium toxicity [97]. We also found that human MTF-1 interacts withp300 and serves as a substrate for acetylation via p300 in vitro. In con-trast to mouse MTF-1, the interaction of human MTF-1 and p300 iszinc independent (VG and WS, unpublished). The findings above[93] suggest a positive role of Sp1 in metal induced transcription,but this might not generally be the case since in other studies, Sp1was reported to act as a negative regulator of the expression ofhuman metallothionein 2a [98,99].

The tumor suppressor protein PTEN was shown to bind to theacidic activation domain of MTF-1 and to be important for MTF-1 de-pendent metallothionein expression [100]. This interaction mainlytakes place in the cytoplasm and the protein phosphatase activity ofPTEN is necessary to activate MTF-1. This is consistent with previousobservations showing that MTF-1 activity is regulated by phosphory-lation [101,102].

Besides interacting with transcriptional coactivators, several stud-ies indicate that MTF-1 interacts with other stress responsive tran-scription factors. The hypoxia-inducible factor 1 alpha (HIF-1α) wasshown to act in concert with MTF-1 to activate mousemetallothionein-1 transcription upon hypoxia [103]. Likewise, MTF-1 was shown to promote HIF-1α dependent gene expression, sinceloss of MTF-1 attenuated HIF-1α nuclear accumulation and transcrip-tional activity [29]. However, it remains to be seen whether this is dueto a direct effect of MTF-1 or to perturbation of hypoxia sensing by theelevated glutathione levels in cells lacking MTF-1.

Several metallothionein genes contain AREs (antioxidant responseelements) next to MREs and are activated via the ARE-binding tran-scription factors of the Nrf family [104].

Although metal-responsive promoters can be synergistically acti-vated by heat and heavy metal stress, experimental alterations ofheat shock factor 1 (HSF1) concentrations had no effect on the tran-scriptional activity of MTF-1 [105]. This argues against an involve-ment of HSF1 in MTF-1/MRE dependent gene regulation. However,the HSF1–MTF-1 interaction is relevant in the converse situation:MTF-1 interacts with HSF1 and attenuates HSF1 mediated gene tran-scription [106]. An additional connection between heat and metalstress is given by the fact that cadmium induces the expression of cer-tain heat shock proteins [107].

Another three transcription factors have been reported to interactwith MTF-1 on metallothionein promoters: upstream stimulatory fac-tor 1 (USF1), nuclear factor 1 (NF1) and CCAAT/enhancer-bindingprotein alpha (C/EBPα). USF1 upregulates transcription of mouse

Ref.

ption of the metallothioneinA (MtnA) gene [92]oters of the MtnA, MtnB and MtnD genes; stimulation of transcription [92]ion of DNA binding, potential chromatin modulator [94]

[95]s transcription of mouse MT-1, but inhibits transcription of human MT-2a [93]

[98]of mouse MT-1, but not ZnT1; MTF-1 is acetylated by p300 in vitro [93] [*]s human MT-2a, MT-1a and ZnT1 expression [100]o induce MT-1 in response to hypoxia [103]MT-1 and MT-2 expression in mouse liver [104]ent Hsp70 expression [105]

[106]MT-1 expression in the visceral yolk sac [108]her enhance or inhibit MTF-1 activity at the MT-1 promoter [109]

[110]expression [111]

1422 V. Günther et al. / Biochimica et Biophysica Acta 1823 (2012) 1416–1425

metallothionein 1 in concert with MTF-1 [108]. The role of NF1 re-mains to be clarified. In one study it was shown to repress metal-lothionein 1 expression [109] whereas another one suggested thatNF1 requires MTF-1 for zinc dependent binding to the promoter to ac-tivate transcription [110]. MTF-1 also cooperates with the product ofone of its target genes: C/EBPα, a tumor suppressor protein enrichedin the liver, positively regulates metallothionein 2a expression inhuman hepatoma (HEP3B) cells [111].

5. Concluding remarks/outlook

The role of MTF-1 in metal homeostasis is well established—it canupregulate expression of exporters and scavengers and repress ex-pression of importers in response to heavy metal exposure. Evidencealso points to a role in other stress response pathways, such as hypox-ia and oxidative stress caused by reactive oxygen species, and in neu-roprotection. The interaction partners of MTF-1 also support thenotion that MTF-1 is interconnected with multiple cell stress re-sponse pathways. At the risk of oversimplification one could concludethat each cell stress response is typically coordinated by a major,stress specific transcription factor but also involves to a lesser extentregulators of other types of stress.

With its central role in protecting a cell from diverse stressors it isnot surprising that MTF-1 is implicated in a diverse set of human dis-eases. Regarding cancer, MTF-1 can be expected to have a beneficialrole for the organism by counteracting pro-cancerogenic conditions,notably heavy metal and oxidative stress. However once a tumoremerges, MTF-1 can promote cancer cell survival by enhancing ex-pression of metallothioneins, stimulation of angiogenesis and extra-cellular matrix remodeling [112,113]. In line with this fact MTF-1was shown to be upregulated in some tumor cells [114]. Resistanceof lymphoma cells to gallium nitrate is partially mediated by MTF-1/metallothionein [115]. Furthermore, a polymorphism in the mouseMTF-1 gene was implicated in the development of γ-ray inducedlymphomas [116]. Of particular interest are the findings in Drosophilamodels of neurodegenerative diseases, namely Alzheimer's and Par-kinson's disease [117,118], which are both characterized by increasedoxidative stress and heavy metal sensitivity in affected neurons [119].There, elevated expression of MTF-1 dramatically improves the con-dition of parkin null mutant flies or transgenic flies expressinghuman amyloid beta peptide (Aβ42) [117,118]. Aβ42 is the maincomponent of amyloid plaques found in brains of Alzheimer'spatients.

Although many new findings have widened our view on MTF-1'srole in heavy metal homeostasis and detoxification and in other cellstress responses, major questions remain. How does MTF-1 activatetranscription of most of its target genes but contributes to the repres-sion of others, such as mouse selenoprotein H and Drosophila Ctr1B? Inthe glucocorticoid hormone response, activation or repression areknown to be dependent on the DNA sequence of the glucocorticoidreceptor binding sites [120]. With MTF-1, both positive and negativeresponses seem to be mediated via the same MRE consensus se-quence; the output rather depends on their spatial arrangement[88] which is thought to dictate interactions with coactivators andco-repressors. In the case of the Drosophila copper importer Ctr1B,the strong MRE sites located around position −700 from the tran-scription start might allow MTF-1 to bind even under copper-replete conditions, thus allowing transcription at low copper. Uponheavy metal load the Ctr1B gene is repressed, possibly via interactionof MTF-1 with a co-repressor [88]. The inhibitory C-terminal domainplays an important role because following its deletion MTF-1 indis-criminately activates both metallothionein and Ctr1B expression innormal food. This is remarkable because other mutations of Drosoph-ila MTF-1 typically hamper metallothionein induction by metal with-out affecting Ctr1B expression.

Elusive is also the exact mechanism of metal sensing byMTF-1 andwhether it is always direct or mediated in concert with other pro-teins. For mammalian MTF-1 the zinc fingers and the acidic activationdomain were shown to mediate zinc sensing, whereas the acidic acti-vation domain works as an independent metal-inducible unit at leastin a subset of cell types [52]. The response to other heavy metals, tooxidative stress and to nitric oxide might be an indirect one, mediatedmainly by release of zinc from metallothioneins and possibly othermetal binding proteins [121–123]. However this model is certainlynot covering all aspects of metal induction, and additional mecha-nisms are contributing (this review, and see also [124]).

Metal specificity is another enigma: how does the cell, always inan MTF-1 dependent manner, ensure the differential response of cer-tain genes to one but not the other metal? One clue might come fromthe finding that for two metallothionein promoters which are prefer-entially induced by copper or cadmium (Drosophila MtnA and MtnB,respectively), activation by MTF-1 involves differential recruitmentof cofactors and Mediator components [92].

Also unclear is the role of the nuclear export signal, since at leastwith human MTF-1, export per se is not essential to regulate metal in-ducibility. It might be necessary for full efficiency and/or anotherfunction of MTF-1. Thus, it would be interesting to see whethergenes repressed by MTF-1, like Zip10, are permanently repressed byan MTF-1 with constitutive nuclear localization, for example as a re-sult of a mutated, dysfunctional nuclear export signal.

We are confident that in the years to come, these open questionswill be successfully addressed by the combined efforts of multiplelaboratories.

Acknowledgements

We thank Kurt Steiner for assistance with data base search andDrs. George Hausmann and Oleg Georgiev for critical reading of themanuscript. This work was supported by the Swiss National ScienceFoundation and the Kanton Zürich.

References

[1] M. Margoshes, B.L. Vallee, A cadmium protein from equine kidney cortex, J. Am.Chem. Soc. 79 (1957) 4813–4814.

[2] T. Dalton, R.D. Palmiter, G.K. Andrews, Transcriptional induction of the mousemetallothionein-I gene in hydrogen peroxide-treated Hepa cells involves a com-posite major late transcription factor/antioxidant response element and metalresponse promoter elements, Nucleic Acids Res. 22 (1994) 5016–5023.

[3] Y. Ren, A. Smith, Mechanism of metallothionein gene regulation byheme-hemopexin. Roles of protein kinase C, reactive oxygen species, andcis-acting elements, J. Biol. Chem. 270 (1995) 23988–23995.

[4] E.J. Kelly, E.P. Sandgren, R.L. Brinster, R.D. Palmiter, A pair of adjacent glucocor-ticoid response elements regulate expression of two mouse metallothioneingenes, Proc. Natl. Acad. Sci. U. S. A. 94 (1997) 10045–10050.

[5] P.F. Searle, G.W. Stuart, R.D. Palmiter, Building a metal-responsive promoterwith synthetic regulatory elements, Mol. Cell. Biol. 5 (1985) 1480–1489.

[6] G.W. Stuart, P.F. Searle, H.Y. Chen, R.L. Brinster, R.D. Palmiter, A 12-base-pairDNA motif that is repeated several times in metallothionein gene promotersconfers metal regulation to a heterologous gene, Proc. Natl. Acad. Sci. U. S. A.81 (1984) 7318–7322.

[7] G.W. Stuart, P.F. Searle, R.D. Palmiter, Identification of multiple metal regulatoryelements in mouse metallothionein-I promoter by assaying synthetic se-quences, Nature 317 (1985) 828–831.

[8] G. Westin, W. Schaffner, A zinc-responsive factor interacts with ametal-regulated enhancer element (MRE) of the mouse metallothionein-Igene, EMBO J. 7 (1988) 3763–3770.

[9] C. Séguin, J. Prévost, Detection of a nuclear protein that interacts with a metalregulatory element of the mouse metallothionein 1 gene, Nucleic Acids Res. 16(1988) 10547–10560.

[10] F. Radtke, R. Heuchel, O. Georgiev, M. Hergersberg, M. Gariglio, Z. Dembic, et al.,Cloned transcription factor MTF-1 activates the mouse metallothionein I pro-moter, EMBO J. 12 (1993) 1355–1362.

[11] E. Brugnera, O. Georgiev, F. Radtke, R. Heuchel, E. Baker, G.R. Sutherland, et al.,Cloning, chromosomal mapping and characterization of the humanmetal-regulatory transcription factor MTF-1, Nucleic Acids Res. 22 (1994)3167–3173.

[12] F. Otsuka, I. Okugaito, M. Ohsawa, A. Iwamatsu, K. Suzuki, S. Koizumi, Novel re-sponses of ZRF, a variant of human MTF-1, to in vivo treatment with heavymetals, Biochim. Biophys. Acta 1492 (2000) 330–340.

1423V. Günther et al. / Biochimica et Biophysica Acta 1823 (2012) 1416–1425

[13] C. Inouye, P. Remondelli, M. Karin, S. Elledge, Isolation of a cDNA encoding ametal response element binding protein using a novel expression cloning proce-dure: the one hybrid system, DNA Cell Biol. 13 (1994) 731–742.

[14] P. Lichtlen, W. Schaffner, Putting its fingers on stressful situations: the heavymetal-regulatory transcription factor MTF-1, Bioessays 23 (2001) 1010–1017.

[15] G.K. Andrews, Cellular zinc sensors: MTF-1 regulation of gene expression, Bio-metals 14 (2001) 223–237.

[16] K. Balamurugan, W. Schaffner, Copper homeostasis in eukaryotes: teetering on atightrope, Biochim. Biophys. Acta. 1763 (2006) 737–746.

[17] D.P. Giedroc, X. Chen, J.L. Apuy, Metal response element (MRE)-binding tran-scription factor-1 (MTF-1): structure, function, and regulation, Antioxid. RedoxSignal. 3 (2001) 577–596.

[18] J.H. Laity, G.K. Andrews, Understanding the mechanisms of zinc-sensing bymetal-response element binding transcription factor-1 (MTF-1), Arch. Biochem.Biophys. 463 (2007) 201–210.

[19] A. Auf der Maur, T. Belser, G. Elgar, O. Georgiev, W. Schaffner, Characterization ofthe transcription factor MTF-1 from the Japanese pufferfish (Fugu rubripes) re-veals evolutionary conservation of heavy metal stress response, Biol. Chem.380 (1999) 175–185.

[20] T.P. Dalton, W.A. Solis, D.W. Nebert, M.J. Carvan III, Characterization of theMTF-1 transcription factor from zebrafish and trout cells, Comp. Biochem. Phy-siol. B Biochem. Mol. Biol. 126 (2000) 325–335.

[21] B. Zhang, D. Egli, O. Georgiev, W. Schaffner, The Drosophila homolog of mamma-lian zinc finger factor MTF-1 activates transcription in response to heavy metals,Mol. Cell. Biol. 21 (2001) 4505–4514.

[22] W.-Y. Chen, J.A.C. John, C.-H. Lin, C.-Y. Chang, Molecular cloning and develop-mental expression of zinc finger transcription factor MTF-1 gene in zebrafish,Danio rerio, Biochem. Biophys. Res. Commun. 291 (2002) 798–805.

[23] A.P.-L. Cheung, C.Y.-M. Au, W.W.-L. Chan, K.M. Chan, Characterization and local-ization of metal-responsive-element-binding transcription factors from tilapia,Aquat. Toxicol. 99 (2010) 42–55.

[24] U. Lindert, L. Leuzinger, K. Steiner, O. Georgiev, W. Schaffner, Characterization ofmetal-responsive transcription factor (MTF-1) from the giant rodent capybarareveals features in common with human as well as with small rodents(mouse, rat). Short communication, Chem. Biodivers. 5 (2008) 1485–1494.

[25] J.C. Rutherford, A.J. Bird, Metal-responsive transcription factors that regulateiron, zinc, and copper homeostasis in eukaryotic cells, Eukaryot. Cell 3 (2004)1–13.

[26] L.H. Moilanen, T. Fukushige, J.H. Freedman, Regulation of metallothionein genetranscription. Identification of upstream regulatory elements and transcriptionfactors responsible for cell-specific expression of the metallothionein genesfrom Caenorhabditis elegans, J. Biol. Chem. 274 (1999) 29655–29665.

[27] C.J. Green, P. Lichtlen, N.T. Huynh, M. Yanovsky, K.R. Laderoute, W. Schaffner,et al., Placenta growth factor gene expression is induced by hypoxia in fibro-blasts: a central role for metal transcription factor-1, Cancer Res. 61 (2001)2696–2703.

[28] B.J. Murphy, G.K. Andrews, D. Bittel, D.J. Discher, J. McCue, C.J. Green, et al., Ac-tivation of metallothionein gene expression by hypoxia involves metal re-sponse elements and metal transcription factor-1, Cancer Res. 59 (1999)1315–1322.

[29] B.J. Murphy, B.G. Sato, T.P. Dalton, K.R. Laderoute, The metal-responsive tran-scription factor-1 contributes to HIF-1 activation during hypoxic stress, Bio-chem. Biophys. Res. Commun. 337 (2005) 860–867.

[30] T.P. Dalton, Q. Li, D. Bittel, L. Liang, G.K. Andrews, Oxidative stress activatesmetal-responsive transcription factor-1 binding activity. Occupancy in vivo ofmetal response elements in the metallothionein-I gene promoter, J. Biol.Chem. 271 (1996) 26233–26241.

[31] G.K. Andrews, Regulation of metallothionein gene expression by oxidative stressand metal ions, Biochem. Pharmacol. 59 (2000) 95–104.

[32] S. Bahadorani, S. Mukai, D. Egli, A.J. Hilliker, Overexpression of metal-responsivetranscription factor (MTF-1) in Drosophila melanogaster ameliorates life-spanreductions associated with oxidative stress and metal toxicity, Neurobiol.Aging 31 (2010) 1215–1226.

[33] C. Günes, R. Heuchel, O. Georgiev, K.H. Müller, P. Lichtlen, H. Blüthmann, et al.,Embryonic lethality and liver degeneration in mice lacking the metal-responsivetranscriptional activator MTF-1, EMBO J. 17 (1998) 2846–2854.

[34] Y. Wang, U. Wimmer, P. Lichtlen, D. Inderbitzin, B. Stieger, P.J. Meier, et al.,Metal-responsive transcription factor-1 (MTF-1) is essential for embryonicliver development and heavy metal detoxification in the adult liver, FASEB J.18 (2004) 1071–1079.

[35] D. Egli, A. Selvaraj, H. Yepiskoposyan, B. Zhang, E. Hafen, O. Georgiev, et al.,Knockout of “metal-responsive transcription factor” MTF-1 in Drosophila by ho-mologous recombination reveals its central role in heavy metal homeostasis,EMBO J. 22 (2003) 100–108.

[36] M.J. Chung, C. Hogstrand, S.-J. Lee, Cytotoxicity of nitric oxide is alleviated byzinc-mediated expression of antioxidant genes, Exp. Biol. Med. 231 (2006)1555–1563.

[37] E.A. Ostrakhovitch, P.-E. Olsson, J. von Hofsten, M.G. Cherian, P53 mediated reg-ulation of metallothionein transcription in breast cancer cells, J. Cell. Biochem.102 (2007) 1571–1583.

[38] M. Hasumi, K. Suzuki, H. Matsui, H. Koike, K. Ito, H. Yamanaka, Regulation ofmetallothionein and zinc transporter expression in human prostate cancercells and tissues, Cancer Lett. 200 (2003) 187–195.

[39] A. Auf der Maur, T. Belser, Y. Wang, C. Günes, P. Lichtlen, O. Georgiev, et al., Char-acterization of the mouse gene for the heavy metal-responsive transcription fac-tor MTF-1, Cell Stress Chaperones 5 (2000) 196–206.

[40] N. Saydam, O. Georgiev, M.Y. Nakano, U.F. Greber, W. Schaffner, Nucleo-cyto-plasmic trafficking of metal-regulatory transcription factor 1 is regulated by di-verse stress signals, J. Biol. Chem. 276 (2001) 25487–25495.

[41] I.V. Smirnova, D.C. Bittel, R. Ravindra, H. Jiang, G.K. Andrews, Zinc and cadmiumcan promote rapid nuclear translocation of metal response element-bindingtranscription factor-1, J. Biol. Chem. 275 (2000) 9377–9384.

[42] R. Heuchel, F. Radtke, O. Georgiev, G. Stark, M. Aguet, W. Schaffner, The tran-scription factor MTF-1 is essential for basal and heavy metal-induced metal-lothionein gene expression, EMBO J. 13 (1994) 2870–2875.

[43] A.L. Guerrerio, J.M. Berg, Metal ion affinities of the zinc finger domains of themetal responsive element-binding transcription factor-1 (MTF1), Biochemistry43 (2004) 5437–5444.

[44] B.M. Potter, L.S. Feng, P. Parasuram, V.A. Matskevich, J.A. Wilson, G.K. Andrews,et al., The six zinc fingers of metal-responsive element binding transcriptionfactor-1 form stable and quasi-ordered structures with relatively small differ-ences in zinc affinities, J. Biol. Chem. 280 (2005) 28529–28540.

[45] J.L. Apuy, X. Chen, D.H. Russell, T.O. Baldwin, D.P. Giedroc, Ratiometric pulsedalkylation/mass spectrometry of the cysteine pairs in individual zinc fingers ofMRE-binding transcription factor-1 (MTF-1) as a probe of zinc chelate stability,Biochemistry 40 (2001) 15164–15175.

[46] X. Chen, M. Chu, D.P. Giedroc, MRE-binding transcription factor-1: weakzinc-binding finger domains 5 and 6 modulate the structure, affinity, and spec-ificity of the metal-response element complex, Biochemistry 38 (1999)12915–12925.

[47] D.C. Bittel, I.V. Smirnova, G.K. Andrews, Functional heterogeneity in the zinc fin-gers of metalloregulatory protein metal response element-binding transcriptionfactor-1, J. Biol. Chem. 275 (2000) 37194–37201.

[48] J.H. Laity, H.J. Dyson, P.E. Wright, DNA-induced alpha-helix capping in conservedlinker sequences is a determinant of binding affinity in Cys(2)-His(2) zinc fin-gers, J. Mol. Biol. 295 (2000) 719–727.

[49] Y. Li, T. Kimura, J.H. Laity, G.K. Andrews, The zinc-sensing mechanism of mouseMTF-1 involves linker peptides between the zinc fingers, Mol. Cell. Biol. 26(2006) 5580–5587.

[50] K. Seipel, O. Georgiev, W. Schaffner, Different activation domains stimulate tran-scription from remote (‘enhancer’) and proximal (‘promoter’) positions, EMBO J.11 (1992) 4961–4968.

[51] F. Radtke, O. Georgiev, H.P. Müller, E. Brugnera, W. Schaffner, Functional do-mains of the heavy metal-responsive transcription regulator MTF-1, NucleicAcids Res. 23 (1995) 2277–2286.

[52] U. Lindert, M. Cramer, M. Meuli, O. Georgiev, W. Schaffner, Metal-responsivetranscription factor 1 (MTF-1) activity is regulated by a nonconventional nucle-ar localization signal and a metal-responsive transactivation domain, Mol. Cell.Biol. 29 (2009) 6283–6293.

[53] H.P. Müller, E. Brungnera, O. Georgiev, M. Badzong, K.H. Müller, W. Schaffner,Analysis of the heavy metal-responsive transcription factor MTF-1 fromhuman and mouse, Somat. Cell Mol. Genet. 21 (1995) 289–297.

[54] X. Chen, B. Zhang, P.M. Harmon, W. Schaffner, D.O. Peterson, D.P. Giedroc, Anovel cysteine cluster in human metal-responsive transcription factor 1 is re-quired for heavy metal-induced transcriptional activation in vivo, J. Biol.Chem. 279 (2004) 4515–4522.

[55] X. He, Q. Ma, Induction of metallothionein I by arsenic via metal-activated tran-scription factor 1: critical role of C-terminal cysteine residues in arsenic sensing,J. Biol. Chem. 284 (2009) 12609–12621.

[56] V. Günther, A.M. Davis, O. Georgiev, W. Schaffner, A conserved cysteine cluster,essential for transcriptional activity, mediates homodimerization of humanmetal-responsive transcription factor-1 (MTF-1), Biochim. Biophys. Acta 1823(2011) 476–483.

[57] Y.-C. Liu, M.-C. Lin, H.-C. Chen, M.F. Tam, L.-Y. Lin, The role of small ubiquitin-likemodifier-interacting motif in the assembly and regulation of metal-responsivetranscription factor 1, J. Biol. Chem. 286 (2011) 42818–42829.

[58] V. Günther, D. Waldvogel, M. Nosswitz, O. Georgiev, W. Schaffner, Dissection ofDrosophila MTF-1 reveals a domain for differential target gene activation uponcopper overload vs. copper starvation, Int. J. Biochem. Cell Biol. 44 (2012)404–411.

[59] X. Chen, H. Hua, K. Balamurugan, X. Kong, L. Zhang, G.N. George, et al., Coppersensing function of Drosophila metal-responsive transcription factor-1 is medi-ated by a tetranuclear Cu(I) cluster, Nucleic Acids Res. 36 (2008) 3128–3138.

[60] B. Bodenmiller, D. Campbell, B. Gerrits, H. Lam, M. Jovanovic, P. Picotti, et al.,PhosphoPep—a database of protein phosphorylation sites in model organisms,Nat. Biotechnol. 26 (2008) 1339–1340.

[61] L. Guo, L.A. Lichten, M.-S. Ryu, J.P. Liuzzi, F.Wang, R.J. Cousins, STAT5-glucocorticoidreceptor interaction and MTF-1 regulate the expression of ZnT2 (Slc30a2) in pan-creatic acinar cells, Proc. Natl. Acad. Sci. U. S. A. 107 (2010) 2818–2823.

[62] S.J. Langmade, R. Ravindra, P.J. Daniels, G.K. Andrews, The transcription factorMTF-1 mediates metal regulation of the mouse ZnT1 gene, J. Biol. Chem. 275(2000) 34803–34809.

[63] U. Wimmer, Y. Wang, O. Georgiev, W. Schaffner, Two major branches ofanti-cadmium defense in the mouse: MTF-1/metallothioneins and glutathione,Nucleic Acids Res. 33 (2005) 5715–5727.

[64] P. Kaler, R. Prasad, Molecular cloning and functional characterization of novelzinc transporter rZip10 (Slc39a10) involved in zinc uptake across rat renalbrush-border membrane, Am. J. Physiol. Renal Physiol. 292 (2007)F217–F229.

[65] L.A. Lichten, M.-S. Ryu, L. Guo, J. Embury, R.J. Cousins, MTF-1-mediated repres-sion of the zinc transporter Zip10 is alleviated by zinc restriction, PLoS One 6(2011) e21526.

1424 V. Günther et al. / Biochimica et Biophysica Acta 1823 (2012) 1416–1425

[66] M.-B. Troadec, D.M. Ward, E. Lo, J. Kaplan, I. De Domenico, Induction of FPN1transcription by MTF-1 reveals a role for ferroportin in transition metal efflux,Blood 116 (2010) 4657–4664.

[67] S. Balesaria, B. Ramesh, H. McArdle, H.K. Bayele, S.K.S. Srai, Divalentmetal-dependent regulation of hepcidin expression by MTF-1, FEBS Lett. 584(2010) 719–725.

[68] S.A. Bellingham, L.A. Coleman, C.L. Masters, J. Camakaris, A.F. Hill, Regulation ofprion gene expression by transcription factors SP1 and metal transcriptionfactor-1, J. Biol. Chem. 284 (2009) 1291–1301.

[69] J.D. Wadsworth, G.S. Jackson, A.F. Hill, J. Collinge, Molecular biology of prionpropagation, Curr. Opin. Genet. Dev. 9 (1999) 338–345.

[70] F.H. Ruiz, E. Silva, N.C. Inestrosa, The N-terminal tandem repeat region of humanprion protein reduces copper: role of tryptophan residues, Biochem. Biophys.Res. Commun. 269 (2000) 491–495.

[71] P.C. Pauly, D.A. Harris, Copper stimulates endocytosis of the prion protein, J. Biol.Chem. 273 (1998) 33107–33110.

[72] D.R. Brown, K. Qin, J.W. Herms, A. Madlung, J. Manson, R. Strome, et al., The cel-lular prion protein binds copper in vivo, Nature 390 (1997) 684–687.

[73] P.C. McHugh, J.A. Wright, D.R. Brown, Transcriptional regulation of thebeta-synuclein 5′-promoter metal response element by metal transcriptionfactor-1, PLoS One 6 (2011) e17354.

[74] F. Mori, M. Nishie, M. Yoshimoto, H. Takahashi, K. Wakabayashi, Reciprocal accu-mulation of beta-synuclein in alpha-synuclein lesions in multiple system atro-phy, Neuroreport 14 (2003) 1783–1786.

[75] V.N. Uversky, J. Li, P. Souillac, I.S. Millett, S. Doniach, R. Jakes, et al., Biophysicalproperties of the synucleins and their propensities to fibrillate: inhibition ofalpha-synuclein assembly by beta- and gamma-synucleins, J. Biol. Chem. 277(2002) 11970–11978.

[76] Z.R. Stoytcheva, V. Vladimirov, V. Douet, I. Stoychev, M.J. Berry, Metal transcriptionfactor-1 regulation via MREs in the transcribed regions of selenoprotein H andother metal-responsive genes, Biochim. Biophys. Acta 1800 (2010) 416–424.

[77] P. Lichtlen, Y. Wang, T. Belser, O. Georgiev, U. Certa, R. Sack, et al., Target genesearch for the metal-responsive transcription factor MTF-1, Nucleic Acids Res.29 (2001) 1514–1523.

[78] M. Cramer, I. Nagy, B.J. Murphy, M. Gassmann, M.O. Hottiger, O. Georgiev, et al.,NF-kappaB contributes to transcription of placenta growth factor and interactswith metal responsive transcription factor-1 in hypoxic human cells, Biol.Chem. 386 (2005) 865–872.

[79] H. Cangul, Hypoxia upregulates the expression of the NDRG1 gene leading to itsoverexpression in various human cancers, BMC Genet. 5 (2004) 27.

[80] H. Ding, B. Wu, H. Wang, Z. Lu, J. Yan, X. Wang, et al., A novel loss-of-functionDDAH1 promoter polymorphism is associated with increased susceptibility tothrombosis stroke and coronary heart disease, Circ. Res. 106 (2010) 1145–1152.

[81] B. Kindermann, F. Döring, J. Budczies, H. Daniel, Zinc-sensitive genes as potentialnew target genes of the metal transcription factor-1 (MTF-1), Biochem. Cell Biol.83 (2005) 221–229.

[82] H. Yepiskoposyan, D. Egli, T. Fergestad, A. Selvaraj, C. Treiber, G. Multhaup, et al.,Transcriptome response to heavymetal stress inDrosophila reveals a new zinc trans-porter that confers resistance to zinc, Nucleic Acids Res. 34 (2006) 4866–4877.

[83] L. Atanesyan, V. Günther, S.E. Celniker, O. Georgiev, W. Schaffner, Characteriza-tion of MtnE, the fifth metallothionein member in Drosophila, J. Biol. Inorg.Chem. 16 (2011) 1047–1056.

[84] D. Egli, J. Domènech, A. Selvaraj, K. Balamurugan, H. Hua, M. Capdevila, et al., Thefour members of the Drosophilametallothionein family exhibit distinct yet over-lapping roles in heavy metal homeostasis and detoxification, Genes Cells 11(2006) 647–658.

[85] J.G. Joshi, A. Zimmerman, Ferritin: an expanded role in metabolic regulation,Toxicology 48 (1988) 21–29.

[86] R. Burke, E. Commons, J. Camakaris, Expression and localisation of the essentialcopper transporter DmATP7 in Drosophila neuronal and intestinal tissues, Int. J.Biochem. Cell Biol. 40 (2008) 1850–1860.

[87] P. de Bie, P. Muller, C. Wijmenga, L.W.J. Klomp, Molecular pathogenesis of Wil-son and Menkes disease: correlation of mutations with molecular defects anddisease phenotypes, J. Med. Genet. 44 (2007) 673–688.

[88] A. Selvaraj, K. Balamurugan, H. Yepiskoposyan, H. Zhou, D. Egli, O. Georgiev,et al., Metal-responsive transcription factor (MTF-1) handles both extremes,copper load and copper starvation, by activating different genes, Genes Dev.19 (2005) 891–896.

[89] K. Balamurugan, D. Egli, H. Hua, R. Rajaram, G. Seisenbacher, O. Georgiev, et al.,Copper homeostasis in Drosophila by complex interplay of import, storage andbehavioral avoidance, EMBO J. 26 (2007) 1035–1044.

[90] A. Gupta, S. Lutsenko, Human copper transporters: mechanism, role in humandiseases and therapeutic potential, Future Med. Chem. 1 (2009) 1125–1142.

[91] B.L. Mihelich, E.A. Khramtsova, N. Arva, A. Vaishnav, D.N. Johnson, A.A. Gian-greco, et al., miR-183–96–182 cluster is overexpressed in prostate tissue andregulates zinc homeostasis in prostate cells, J. Biol. Chem. 286 (2011)44503–44511.

[92] M.T. Marr II, Y. Isogai, K.J. Wright, R. Tjian, Coactivator cross-talk specifies tran-scriptional output, Genes Dev. 20 (2006) 1458–1469.

[93] Y. Li, T. Kimura, R.W. Huyck, J.H. Laity, G.K. Andrews, Zinc-induced formation of acoactivator complex containing the zinc-sensing transcription factor MTF-1,p300/CBP, and Sp1, Mol. Cell. Biol. 28 (2008) 4275–4284.

[94] L. Giot, J.S. Bader, C. Brouwer, A. Chaudhuri, B. Kuang, Y. Li, et al., A protein inter-action map of Drosophila melanogaster, Science 302 (2003) 1727–1736.

[95] A. Vardanyan, L. Atanesyan, D. Egli, S.J. Raja, M. Steinmann-Zwicky, R. Renkawitz--Pohl, et al., Dumpy-30 family members as determinants of male fertility and

interaction partners ofmetal-responsive transcription factor 1 (MTF-1) inDrosoph-ila, BMC Dev. Biol. 8 (2008) 68.

[96] F. Okumura, Y. Li, N. Itoh, T. Nakanishi, M. Isobe, G.K. Andrews, et al., Thezinc-sensing transcription factor MTF-1 mediates zinc-induced epigeneticchanges in chromatin of the mouse metallothionein-I promoter, Biochim. Bio-phys. Acta 1809 (2011) 56–62.

[97] T. Kimura, Y. Li, F. Okumura, N. Itoh, T. Nakanishi, T. Sone, et al., Chromium(VI)inhibits mouse metallothionein-I gene transcription by preventing thezinc-dependent formation of an MTF-1–p300 complex, Biochem. J. 415 (2008)477–482.

[98] S. Koizumi, F. Otsuka, Nuclear proteins binding to the humanmetallothionein-IIA gene upstream sequences, Ind. Health 32 (1994) 193–205.

[99] Y. Ogra, K. Suzuki, P. Gong, F. Otsuka, S. Koizumi, Negative regulatory role of Sp1in metal responsive element-mediated transcriptional activation, J. Biol. Chem.276 (2001) 16534–16539.

[100] M.-C. Lin, Y.-C. Liu, M.F. Tam, Y.-J. Lu, Y.-T. Hsieh, L.-Y. Lin, PTEN interacts withmetal-responsive transcription factor 1 and stimulates its transcriptional activ-ity, Biochem. J. 441 (2012) 367–377.

[101] N. Saydam, T.K. Adams, F. Steiner, W. Schaffner, J.H. Freedman, Regulation ofmetallothionein transcription by the metal-responsive transcription factorMTF-1: identification of signal transduction cascades that controlmetal-inducible transcription, J. Biol. Chem. 277 (2002) 20438–20445.

[102] O. LaRochelle, V. Gagné, J. Charron, J.W. Soh, C. Séguin, Phosphorylation is in-volved in the activation of metal-regulatory transcription factor 1 in responseto metal ions, J. Biol. Chem. 276 (2001) 41879–41888.

[103] B.J. Murphy, T. Kimura, B.G. Sato, Y. Shi, G.K. Andrews, Metallothionein inductionby hypoxia involves cooperative interactions between metal-responsive tran-scription factor-1 and hypoxia-inducible transcription factor-1alpha, Mol. Can-cer Res. 6 (2008) 483–490.

[104] M. Ohtsuji, F. Katsuoka, A. Kobayashi, H. Aburatani, J.D. Hayes, M. Yamamoto,Nrf1 and Nrf2 play distinct roles in activation of antioxidant responseelement-dependent genes, J. Biol. Chem. 283 (2008) 33554–33562.

[105] N. Saydam, F. Steiner, O. Georgiev, W. Schaffner, Heat and heavy metal stresssynergize to mediate transcriptional hyperactivation by metal-responsive tran-scription factor MTF-1, J. Biol. Chem. 278 (2003) 31879–31883.

[106] R. Uenishi, P. Gong, K. Suzuki, S. Koizumi, Cross talk of heat shock and heavy metalregulatory pathways, Biochem. Biophys. Res. Commun. 341 (2006) 1072–1077.

[107] J.H. Ovelgönne, J.E. Souren, F.A. Wiegant, R. Van Wijk, Relationship betweencadmium-induced expression of heatshock genes, inhibition of protein synthe-sis and cell death, Toxicology 99 (1995) 19–30.

[108] G.K. Andrews, D.K. Lee, R. Ravindra, P. Lichtlen, M. Sirito, M. Sawadogo, et al., Thetranscription factors MTF-1 and USF1 cooperate to regulate mousemetallothionein-I expression in response to the essential metal zinc in visceralendoderm cells during early development, EMBO J. 20 (2001) 1114–1122.

[109] S. Majumder, K. Ghoshal, R.M. Gronostajski, S.T. Jacob, Downregulation of con-stitutive and heavy metal-induced metallothionein-I expression by nuclear fac-tor I, Gene Expr. 9 (2001) 203–215.

[110] O. LaRochelle, S. Labbé, J.-F. Harrisson, C. Simard, V. Tremblay, G. St-Gelais, et al.,Nuclear factor-1 and metal transcription factor-1 synergistically activate themouse metallothionein-1 gene in response to metal ions, J. Biol. Chem. 283(2008) 8190–8201.

[111] J. Datta, S. Majumder, H. Kutay, T. Motiwala, W. Frankel, R. Costa, et al., Metal-lothionein expression is suppressed in primary human hepatocellular carcino-mas and is mediated through inactivation of CCAAT/enhancer binding proteinalpha by phosphatidylinositol 3-kinase signaling cascade, Cancer Res. 67(2007) 2736–2746.

[112] Z.A. Haroon, K. Amin, P. Lichtlen, B. Sato, N.T. Huynh, Z. Wang, et al., Loss ofmetal transcription factor-1 suppresses tumor growth through enhanced matrixdeposition, FASEB J. 18 (2004) 1176–1184.

[113] B.J. Murphy, Regulation of malignant progression by the hypoxia-sensitive tran-scription factors HIF-1α and MTF-1, Comp. Biochem. Physiol. B Biochem. Mol.Biol. 139 (2004) 495–507.

[114] Y. Shi, K. Amin, B.G. Sato, S.J. Samuelsson, L. Sambucetti, Z.A. Haroon, et al., Themetal-responsive transcription factor-1 protein is elevated in human tumors,Cancer Biol. Ther. 9 (2010) 469–476.

[115] M. Yang, S.H. Kroft, C.R. Chitambar, Gene expression analysis of gallium-resistantand gallium-sensitive lymphoma cells reveals a role formetal-responsive transcrip-tion factor-1, metallothionein-2A, and zinc transporter-1 in modulating the anti-neoplastic activity of gallium nitrate, Mol. Cancer Ther. 6 (2007) 633–643.

[116] M. Maruyama, T. Yamamoto, Y. Kohara, Y. Katsuragi, Y. Mishima, Y. Aoyagi, et al.,Mtf-1 lymphoma-susceptibility locus affects retention of large thymocytes withhigh ROS levels in mice after gamma-irradiation, Biochem. Biophys. Res. Com-mun. 354 (2007) 209–215.

[117] H. Hua, L. Münter, A. Harmeier, O. Georgiev, G. Multhaup, W. Schaffner, Toxicity ofAlzheimer's disease-associated Aβ peptide is ameliorated in a Drosophilamodel bytight control of zinc and copper availability, Biol. Chem. 392 (2011) 919–926.

[118] N. Saini, O. Georgiev, W. Schaffner, The parkin mutant phenotype in the fly islargely rescued by metal-responsive transcription factor (MTF-1), Mol. Cell.Biol. 31 (2011) 2151–2161.

[119] K. Jomova, D. Vondrakova, M. Lawson, M. Valko, Metals, oxidative stress andneurodegenerative disorders, Mol. Cell. Biochem. 345 (2010) 91–104.

[120] A. Dostert, T. Heinzel, Negative glucocorticoid receptor response elements andtheir role in glucocorticoid action, Curr. Pharm. Des. 10 (2004) 2807–2816.

[121] M.M. Cortese-Krott, M. Münchow, E. Pirev, F. Hessner, A. Bozkurt, P. Uciechowski,et al., Silver ions induce oxidative stress and intracellular zinc release in humanskin fibroblasts, Free Radic. Biol. Med. 47 (2009) 1570–1577.

1425V. Günther et al. / Biochimica et Biophysica Acta 1823 (2012) 1416–1425

[122] M.S. Stitt, K.J. Wasserloos, X. Tang, X. Liu, B.R. Pitt, C.M. St Croix, Nitricoxide-induced nuclear translocation of the metal responsive transcription fac-tor, MTF-1 is mediated by zinc release from metallothionein, Vascul. Pharmacol.44 (2006) 149–155.

[123] B. Zhang, O. Georgiev, M. Hagmann, C. Günes, M. Cramer, P. Faller, et al., Activityof metal-responsive transcription factor 1 by toxic heavy metals and H2O2 invitro is modulated by metallothionein, Mol. Cell. Biol. 23 (2003) 8471–8485.

[124] J.-P. Bourdineaud, M. Baudrimont, P. Gonzalez, J.-L. Moreau, Challenging the modelfor induction ofmetallothionein gene expression, Biochimie 88 (2006) 1787–1792.

[125] G.E. Crooks, G. Hon, J.-M. Chandonia, S.E. Brenner, WebLogo: a sequence logogenerator, Genome Res. 14 (2004) 1188–1190.

[126] Y. Wang, I. Lorenzi, O. Georgiev, W. Schaffner, Metal-responsive transcriptionfactor-1 (MTF-1) selects different types of metal response elements at low vs.high zinc concentration, Biol. Chem. 385 (2004) 623–632.