Mechanisms through which gene expression is regulated by zinc are central to cellular zinc homoeostasis. In this context, evidence for the involvement of zinc dyshomoeostasis in the aetiology of diseases, including Type 2 diabetes, Alzheimer's disease and cancer, highlights the importance of zinc-regulated gene expression. Mechanisms elucidated in bacteria and yeast provide examples of different possible modes of zinc-sensitive gene regulation, involving the zinc-regulated binding of transcriptional activators and repressors to gene promoter regions. A mammalian transcriptional regulatory mechanism that mediates zinc-induced transcriptional up-regulation, involving the transcription factor MTF1 (metal-response element-binding transcription factor 1), has been studied extensively. Gene responses in the opposite direction (reduced mRNA levels in response to increased zinc availability) have been observed in mammalian cells, but a specific transcriptional regulatory process responsible for such a response has yet to be identified. Examples of single zinc-sensitive transcription factors regulating gene expression in opposite directions are emerging. Although zinc-induced transcriptional repression by MTF1 is a possible explanation in some specific instances, such a mechanism cannot account for repression by zinc of all mammalian genes that show this mode of regulation, indicating the existence of as yet uncharacterized mechanisms of zinc-regulated transcription in mammalian cells. In addition, recent findings reveal a role for effects of zinc on mRNA stability in the regulation of specific zinc transporters. Our studies on the regulation of the human gene SLC30A5 (solute carrier 30A5), which codes for the zinc transporter ZnT5, have revealed that this gene provides a model system by which to study both zinc-induced transcriptional down-regulation and zinc-regulated mRNA stabilization.
- metal-response element-binding transcription factor 1 (MTF1)
- mRNA stability
- solute carrier 30A5 gene (SLC30A5)
- zinc transporter 5 (ZnT5)
Genome analysis indicates that between 3 and 10% of all human genes may code for proteins that bind zinc . The redox stability of zinc (contrasting with other transition metals) coupled with its ability to form polyhedral co-ordination complexes with a variety of ligands, notably histidine and cysteine, renders zinc a very useful component of cellular proteins where it may play a structural role, typified by the zinc-finger domains of DNA-binding proteins such as transcription factors, or be involved in enzyme catalysis. All six major enzyme classes contain examples of zinc-containing proteins. These diverse and prevalent roles of zinc in biology highlight the importance of mechanisms to maintain zinc homoeostasis, at both the whole organism and cellular levels. Recent research reveals a central role for zinc dyshomoeoastasis in a number of important non-communicable disease processes. Notably, human genome association studies reveal a strong association of a specific allele of the SLC30A8 (solute carrier 30A8) gene (zinc transporter ZnT8) with Type 2 diabetes mellitus [2–4], and results of recent functional studies corroborate this association [5,6]. A role for zinc in the pathology of Alzheimer's disease has been longer-established, supported by many lines of evidence, including the accumulation of zinc in Alzheimer's disease cortex  and the observation that perturbation of zinc homoeostasis, either pharmacologically  or through diet , affects disease symptoms in mouse models of the disease. Cellular zinc dyshomoeostasis is also implicated in the progression of various cancers, notably prostate cancer  and also breast cancer .
Cellular and whole-body zinc homoeostasis is therefore of fundamental importance and, in mammalian systems, is maintained through mechanisms that include regulation of the expression of genes involved in zinc homoeostatic pathways. These genes include those coding for transporters responsible for the flux of zinc across the plasma membrane and intracellular membranes and for the intracellular zinc-binding protein metallothionein. Several published studies employing transcriptomic-based approaches, such as microarray hybridization, have identified many genes regulated by increased or reduced zinc availability in both a positive a negative manner, as well as genes with more complex response profiles to zinc [12–21].
Fundamental mechanisms of zinc-regulated gene transcription
Theoretically, four fundamental mechanisms based on the binding of protein regulatory factors to gene promoter regions may underlie transcriptional regulation by zinc, differentiated by whether the protein factor when bound exerts a positive or negative influence on transcription and whether the protein factor shows increased or reduced binding to the promoter region under conditions of increased zinc availability (Figure 1). One must look beyond the mammals to identify an example of each mechanism. An example of transcriptional activation through binding of a factor under conditions of reduced zinc availability (type A mechanism, as represented in Figure 1) is provided by the Saccharomyces cerevisiae transcriptional activator Zap1, which binds to an 11 bp ZRE (zinc-responsive element)  in the promoter regions of genes including the high-affinity zinc-uptake transporter ZRT1  and the vacuolar zinc exporter ZRT3 . The prokaryotic SmtB/ArsR family provides an example of negative transcription regulatory factors that bind to their target promoter regions under conditions of reduced zinc availability (type B mechanism, as represented in Figure 1). These proteins dissociate with their binding site on DNA under conditions of increased intracellular zinc concentration to relieve transcriptional repression (reviewed in ). The Escherichia coli transcriptional inhibitory factor Zur, which controls expression of the znuABC operon, coding for an ABC family zinc uptake transporter, binds as a homodimer under conditions of increased intracellular zinc concentration to a regulatory sequence in the operator region , providing an example of transcriptional inhibition through binding of a regulatory factor at higher zinc levels (type C mechanism, as represented in Figure 1). In mammals, the only well-characterized gene regulatory response to cellular zinc status is the increased transcription of genes at higher levels of zinc availability mediated through binding of the transcription factor MFT1 [MRE (metal-response element)-binding transcription factor 1] to copies of the MRE consensus sequence in the promoter region, providing an example of transcriptional activation through binding of a regulatory factor at higher zinc levels (type D mechanism, as represented in Figure 1). Metallothoinein-I and -II genes and the gene for the zinc transporter ZnT1, responsible for the efflux of zinc across the plasma membrane, are up-regulated transcriptionally through this mechanism [27,28]. Mammalian MTF1 proteins show a high degree of conservation, particularly in the six-Cys2-His2 zinc-finger DNA-binding domain. Studies on aspects of MTF1 function, including the contribution of specific zinc fingers, individually or co-operatively, to cellular zinc sensing and DNA binding (reviewed in ) are advancing our knowledge towards an in-depth understanding of the function of MTF1, both with respect to its role in zinc-regulated gene expression and also in transcriptional regulation in response to a wide variety of environmental stresses in which it also plays an important role.
Complexities of zinc-regulated gene transcription
A number of reports document tissue- or cell-type-specific regulation of gene expression by zinc, including variations in the direction and magnitude of change [30–34], indicating either the action of multiple (possibly tissue-specific) zinc-sensitive regulatory factors on single gene promoter regions or differential sensitivity of specific regulatory factors to zinc availability, depending on the specific cellular environment. An additional level of complexity is invoked by the fact that recent evidence highlights opposing roles of individual regulatory protein factors in the regulation of specific genes. For example, limited evidence indicates a degree of plasticity within both the S. cerevisiae (Zap1) and mammalian (MTF1) systems, specifically regulation by zinc in the direction that opposes the more usual mechanism. Zap1-mediated transcriptional repression, rather than activation, has been observed for the S. cerevisiae ZTR2 gene, coding for the low-affinity zinc uptake transporter, and is through binding to an element downstream of the TATA box . A second example of Zap1-mediated transcriptional repression involves the gene for the major S. cerevisiae zinc-dependent alcohol dehydrogenase, ADH1. Binding of Zap1 to the ADH1 promoter upstream of the binding site of the activator Rap1 induces an intergenic RNA transcript whose expression displaces Rap1, leading to ADH1 repression . Suggestions that MTF1 may also mediate the negative regulation of gene expression in response to zinc in mammalian cells have been based on the observation that expression of the zinc influx transporter Zip10, which includes MREs in the upstream region, was increased in the conditional MTF1-knockout mouse . Recently, a role for an MRE cluster in zinc-induced transcriptional repression of the zebrafish ZIP10 gene was reported . The evidence was based on coupling of this promoter region, and mutations thereof, to a reporter gene and is suggestive of zinc-induced transcriptional repression through MTF1 . Such observations indicate the possibility that MTF1 may mediate both transcriptional activation and repression at higher levels of zinc availability. Mutation of the consensus MRE sequence in a zinc-repressed promoter–reporter construct based on the upstream region of the SLC30A5 gene, coding for the human zinc transporter ZnT5, however, failed to abolish zinc-induced transcriptional repression in transfected human intestinal Caco-2 cells . This observation indicates that a mechanism of transcriptional regulation alternative to MTF1 operates in mammalian cells and highlights the SLC30A5 gene as a model system on which to base studies aimed at elucidating this mode of zinc-sensitive transcriptional regulation.
Effects of zinc on mRNA stability
A complete description of zinc-regulated gene expression also requires that possible effects of zinc on mRNA stability are considered. Indeed, our recent finding that the mRNA for ZnT5 (SLC30A5 gene) is stabilized by increased zinc availability in the human intestinal Caco-2 cell line  identifies that modulation of transcript stability is a mechanism, additional to transcriptional regulation, through which regulated gene expression in response to zinc, and hence zinc homoeostasis, is achieved. A second recent report corroborates the view that zinc-regulated mRNA stability may be a general mechanism through which gene expression in response to zinc is achieved, accounting for increased expression of the mRNA for the mouse zinc uptake protein Zip4 (Slc39a4 gene) in response to zinc deficiency and repression in response to zinc repletion . The regulation by iron of the stability of transcripts involved in iron homoeostasis, exemplified by the transferrin receptor, is a well-characterized example of the regulation of gene expression by a metal through such a mechanism . Iron destabilizes transcripts that include the IRE (iron regulatory element) in the 3′-UTR (untranslated region) by causing the release of cytosolic aconitase, whose binding to the IRE stabilizes the transcript. Generally, the 3′-UTR controls mRNA stability and so is the likely binding site in the ZnT5 and Zip4 mRNAs for a protein factor(s) that has a stabilizing or destabilizing role.
The human SLC30A5 gene: a model system for studies to elucidate novel mechanisms of zinc-responsive gene expression
The above discussion identifies the SLC30A5 gene, and its mRNA transcript (ZnT5), as a model system on which to base studies aimed at elucidating novel features of zinc-regulated gene expression in mammalian systems, specifically zinc-induced transcriptional repression and zinc-regulated mRNA stability. ZnT5 transcripts are detected in the cell as two major splice variants, both transcribed from a common promoter region . Variant A is localized to the Golgi apparatus [39,42], where it appears to be involved in the delivery of zinc to enzymes entering the secretory pathway [43,44], and variant B is expressed at the plasma membrane [31,39], where evidence for bidirectional function  indicates possible roles in both the uptake and efflux of zinc. Regulation of the SLC30A5 gene appears complex, and both increased and reduced expression has been measured in response to zinc in a variety of systems, including increased expression in post-confluent Caco-2 cells exposed to 100 μM zinc for 7 days , reduced expression in human intestinal mucosa in response to a daily supplement , and reduced expression in mouse placenta in response to both a zinc-restricted and zinc-supplemented diet . This apparently complex mode of regulation may reflect the fact that two opposing pathways appear to operate in response to increased zinc availability: transcriptional repression and increased mRNA stability .
Current research in our laboratory on the transcriptional regulation of the SLC30A5 gene is focused on the use of EMSA (electrophoretic mobility-shift analysis) to identify sites for zinc-dependent protein-binding events in the SLC30A5 promoter, corroborated by testing candidate binding sequences which may confer zinc-sensitivity of reporter gene expression. We are also investigating the role of the ZnT5 3′-UTR in zinc-mediated regulation of the expression of a coupled reporter gene, complemented by the investigation of protein binding to this region using EMSA. These studies have the promise to identify novel fundamental mechanisms of zinc-responsive gene regulation in mammalian systems and so to add to understanding of normal cellular zinc homoeostasis and therefore also potentially to improve knowledge about pathologies involving dysregulated zinc homoeostasis.
We acknowledge the financial support of the BBSRC (Biotechnology and Biological Sciences Research Council).
Metal Metabolism: Transport, Development and Neurodegeneration: A Biochemical Society Focused Meeting held at Imperial College London, U.K., 9–10 July 2008. Organized and Edited by David Allsop (Lancaster, U.K.) and Harry McArdle (Rowett Research Institute, Aberdeen, U.K.).
Abbreviations: ADH1, alcohol dehydrogenase 1; EMSA, electrophoretic mobility-shift analysis; IRE, iron regulatory element; MRE, metal-response element; MTF1, MRE-binding transcription factor 1; Slc, solute carrier; UTR, untranslated region; ZnT, zinc transporter; ZRE, zinc-response element
- © The Authors Journal compilation © 2008 Biochemical Society