Deborah B. Zamble, University of Toronto, Ontario, Canada
Metalloregulators are proteins that bind metals and modulate gene expression through direct interactions with DNA or RNA. The genes under this metal-dependent control encode a variety of proteins involved in the cellular homeostasis of both essential and toxic metals. Metalloregulators are present in all types of organisms, and extensive information exists about their mechanisms, although many unanswered questions remain. The global activities of these metal-responsive factors require overcoming complex challenges, such as the manner in which the proteins regulate gene expression, the mechanisms of the metal-dependent protein conformation transformations, and the ability of the proteins to recognize the designated metal(s). An understanding of these key biomolecules draws from disciplines such as cell biology, protein chemistry, and inorganic chemistry and provides molecular insight into one fundamental aspect of life.
Although a variety of biologic processes are regulated by metals, for the purposes of this review metalloregulators are defined as proteins that act as metal-responsive genetic switches (1). These proteins are sensors that monitor the cellular levels of one or more metal ions and then respond to changes in availability by modulating the expression of a variety of metal pathways. A few examples of metalloregulatory pathways exist in which the two responsibilities, sensing and regulating, are divided up between separate protein components. Learning how metalloregulators function entails defining the cell biology of what they do and the bioinorganic and bioorganic chemistry of how they do it. This article considers some of these aspects in general and then addresses them more specifically in the context of examples. The focus is on how these proteins bind and respond to metal ions. Other properties, such as the details of specific protein contacts with nucleic acids or RNA polymerase, will not be discussed. For more information on that area, the reader is referred to the article entitled “Transcription Factors.” Furthermore, the scope of this article includes protein families that respond to transition metals, both essential and poisonous, as well as the environmental toxins arsenic and lead. The alkali and alkaline earth metals will not be included here. Finally, the examples described are predominantly from prokaryotic organisms because these systems are more clearly defined than the eukaryotic versions. Several reviews on eukaryotic regulators are listed at the end of this article under “Further Reading.” Unfortunately, space limitations prohibit citing many of the primary references of the information discussed below, so the references are limited to a small number of publications. The reader is directed to the comprehensive review articles that are listed at the beginning of each section, which contain all of the appropriate references.
Many metal ions that can get into cells, such as mercury or lead, are poisonous and must be neutralized and exported as quickly as possible. However, even metals that have an essential cellular role, such as zinc, iron, or copper, can be toxic in excess. An organism must ensure an adequate supply of these nutrients while keeping the concentrations under tight control to prevent accumulation and cellular damage. Furthermore, each essential metal performs distinct cellular functions and seems to be regulated independently. Consequently, the metalloregulators are not only sensitive to changes in metal availability, caused by variations in the external supply or the nutritional needs of the organism, but they must be able to discriminate between the various types of metals present in the biological system. Metalloregulators contribute to the maintenance of this delicate balance by controlling the expression of metal uptake and export pathways, detoxification and storage/sequestration systems, as well as proteins that employ the metals such as metalloenzymes. The activity of the metalloregulators is linked intimately to the operations of these metal-centered pathways. For more information, see the articles on “Metal Complexes, Assembly of,” “Metal Homeostasis, “intercellular,” “Metal Homeostasis: An Overview,” “Metal Transport through Membranes,” and “Metallochaperones, Chemistry of.”
The most common genetic control point of the metalloregulators is transcription, with one major exception in iron regulation discussed below. The response elicited by the metal can be repression, derepression, and/or activation of transcription (Fig. 1). Proteins are classified into families based on sequence homology (Table 1), and usually they respond in the same manner as other family members to metal ions, although they may differ in metal selectivity. In many cases, the mechanism of genetic control is fairly straightforward. Metal binding to members of the ArsR/SmtB, DtxR, Fur, and NikR families of metalloregulators either activates or inhibits DNA binding. The DNA recognition sequences of these proteins are close to or are overlapping the transcription start sites in the promoters of the genes that they regulate, so it is likely that DNA binding by these proteins (either the apo or holo, depending on the system) sterically blocks transcription initiation by the RNA polymerase (2, 3, 4, 5).
However, as described in this section, the regulation by some other metalloregulators is more complicated. Several examples are discussed, each showcasing a possible mechanism of genetic control. Although the purpose is to highlight the cellular biology of how these systems work, the biological chemistry is an integral component of their activities and thus is a part of the discussion.
Table 1. Families of prokaryotic metalloregulators that function as transcription factors that are
Response to Metal
ArsR SmtB CzrA NmtR CadC
Zn(II)/Co(II) Ni(II)/Co(II) Cd(II)/Pb(II)/Bi(III)/Zn(II)
Release of DNA
Derepression of metal
resistance proteins, efflux transporters, metal storage
MerR ZntR CueR
Activation of metal
resistance proteins, efflux transporters
DtxR IdeR MntR
Fe(II) Fe(II) Mn(II), Cd(II)
Repression of uptake transporters, virulence factors
Fur Zur Nur Mur
Fe(II) Zn(II) Ni(II) Mn(II)
Repression of: uptake transporters, metal scavengers, virulence factors, other cellular functions
Repression of uptake transporters Activation of: nickel enzymes
aThis table is not an exhaustive list of metal-dependent members of these families. Furthermore, proteins in these families that respond to factors other than metals that are not listed here.
Figure 1. Simple models of metalloregulated processes. (a-c) The RNA polymerase (RNAP) binds to the -35 and -10 sequences (gray boxes) in the promoter of the regulated gene (arrow) and initiates transcription. The metal (Me) binds to the metalloregulator (ovals), which may not be a dimer in the absence of metals, and causes a conformational change in the protein that influences transcription. (d) IRE-BP (sphere) binds to the mRNA encoding iron-using or iron uptake gene and prevents translation or degradation, respectively.
One of the first metalloregulatory proteins to be characterized extensively is the prokaryotic MerR transcription factor (1, 6, 7), which acts either as a repressor (apo-protein) or an activator (holo-protein) of the mer operon encoding mercury resistance proteins (Fig. 1c). The -35 and -10 sequence elements of the mer promoter, binding sites for the RNA polymerase initiation complex, are separated by an unusually long distance that results in poor constitutive transcription. Apo-MerR binds to the DNA between these sequences and bends the DNA, which results in a slight increase in repression on the suboptimal promoter. It also recruits the RNA polymerase to the transcription start site where it waits in a stalled complex. Upon binding of Hg(II), MerR undergoes a conformational change that causes the DNA to straighten out and to unwind; these distortions place the -10 and -35 sites in optimal positions for productive initiation of transcription. MerR can activate fully the mer promoter over a small range (less than an order of magnitude) of mercuric salt concentrations that range around 10-8 M in vitro and slightly higher in cell culture experiments. This sensitive and cooperative sensor provides a rapid and robust response that only is turned on when needed and is activated fully before cytoplasmic Hg(II) concentrations reach levels that affect cell growth.
Once the mercury has been eliminated from the cell, it would be wasteful energetically to continue to synthesize the resistance proteins. However, it is unclear how the Hg(II)-MerR complex could disengage from the promoter in a timely manner. It is possible that apo-MerR can displace the metal complex at the promoter because it binds the recognition sequence with slightly higher affinity (7), or that the Hg(II) is released eventually from the MerR-DNA complex. However, a recent study suggests that another protein encoded by the mer operon, MerD, may play a role in switching off the induction (8). MerD, which shares sequence homology with MerR, can form a ternary complex with apo-MerR and the DNA recognition complex. At the addition of mercury, the presence of MerD causes a fraction of the DNA to be released, which allows the expression of the divergently transcribed merR gene. Once mercury is eliminated from the cytosol, then the newly produced apo protein would bind to the empty promoter and repress transcription of the mer operon.
Post-transcriptional iron regulation
In mammals, as well as in certain other species, iron metabolism is regulated at the posttranscriptional level (for reviews see References 9-12). Two homologous iron-regulatory proteins (IRP1 and IRP2) bind with high affinity to specific RNA sequences called iron-responsive elements (IREs), which are present in the untranslated regions of the mRNAs that encode many of the proteins involved in iron metabolism. The IREs are conserved hairpin structures, but subtle differences in the sequences of the IREs and the surrounding mRNA affect IRP binding and fine-tune the strength of the interaction.
The IREs are bound by the IRPs at low iron concentrations (Fig. 1d). Two different effects of IRP binding exist, depending on the location of the IRE in the mRNA. In the case of proteins that would not be useful under limiting iron conditions, such as ferritin (iron storage), ferroportin (iron efflux), and aminolevulinate synthase (heme biosynthesis), the IRE is near the translation start site at the 5' end of the mRNAs and IRP binding blocks translation initiation and protein production. In contrast, under the same iron-deficient conditions, the mRNA for the transferrin receptor (iron uptake) is protected from degradation by IRP binding to multiple IREs at the 3' end of the mRNA, which enhances protein production and leads to an increase in iron intake. In an iron-replete situation the IRPs do not bind to the mRNA so the effects are reversed: The genes for iron-using proteins are translated, and the transferrin receptor mRNA is degraded.
The RNA-binding activities of the two IRPs have different mechanisms of inhibition by iron, and not all of the details are defined clearly. The binding of IRP1 to mRNA is blocked by the formation of an [4Fe-4S] cluster, which allows the protein to function as a cytosolic aconitase enzyme. IRP2 does not have an iron cluster, but it is targeted for proteosomal degradation in an iron-dependent process. The IRPs are regulated by a variety of additional factors such as heme, oxidative stress, nitric oxide, and phosphorylation by intracellular signaling factors, indicating that IRPs provide a bridge between iron metabolism and other cellular pathways.
Other links to translation
Another link to posttranscriptional regulation of metal homeostasis is the global iron regulator Fur (4). Fur controls the transcription of a large number of genes (more than 90 in E. coli), most of which encode proteins involved in iron acquisition as well as other essential metabolic pathways. The Fe(II)-Fur complex represses the transcription of these genes when it binds to a recognition sequence located in between the -35 and -10 sites. A subset of Fur-regulated genes seems to be activated instead of repressed by the iron-Fur complex, but these genes do not have an obvious Fur recognition sequence in the DNA promoters. This mystery was solved by the discovery that Fe(II)-Fur represses the production of a small RNA called RyhB (13). If iron is limiting, the Fur repression is alleviated and RyhB is produced. In turn, RhyB inhibits the expression of the target genes, which encode proteins involved in iron storage or iron-using enzymes, by blocking translation and/or promoting degradation of the mRNAs (13, 14). In this manner, it has been suggested that the production of nonessential iron proteins is shut down quickly and that the limited iron resources can be redirected to critical functions before iron uptake is upregulated and fills the reserves, which requires multiple steps initiated by the derepression of transcription. This activity of Fur results in a concerted positive and negative response to changes in iron levels that employs the same metalloregulator.
Several pathways exist in which the metal sensing and the transcriptional regulatory functions are handled by two separate proteins (6, 15). A sensor protein spans the cytoplasmic membrane, senses the amount of metals in the periplasm, and communicates this information to the regulatory protein that is inside of the cell. When the concentration of the appropriate metal reaches critical levels, the internal kinase domain of the sensor protein phosphorylates the cytosolic regulatory protein, which activates it to control transcription. These two-component systems are thought to help protect the periplasm from damage.
One very unusual arrangement that diverges from the pathway described above is the cop operon, which is involved in copper homeostasis and found in E. hirae and closely related Gram-positive bacteria (16). CopY, a dimeric repressor with significant homology to P-lactamse repressors, binds the cop promoter in a zinc-loaded form. The zinc is displaced by two Cu(I) ions; this dinuclear copper cluster causes the protein to release the cop DNA, which also encodes an efflux transporter. The copper is supplied by the metallochaperone CopZ, which forms a heterodimer with CopY via specific electrostatic interactions. CopZ obtains the copper from the uptake transporter CopA, also by means of directed protein-protein interactions. This type of system ensures that the potentially toxic copper ions always are bound by a protein factor and eliminate any need for unprotected free copper to be available in the cytoplasm as a signal to the genetic regulator. CopZ is homologous to copper chaperones from other species, but it is not clear whether the mechanism of direct copper transfer to the corresponding metalloregulator is conserved in other organisms.
The binding of the metal ion coregulator to a metalloregulator protein is coupled to a protein conformational change such that it alters the DNA or RNA complex. In several examples, high-resolution structural studies have shed some light on the mechanisms of this allosteric response.
In the absence of metal, members of the ArsR/SmtB class of transcription factors bind to their respective DNA promoters and inhibit transcription; metal binding to the proteins decreases the affinity for DNA and allows transcription to proceed (derepression, Fig. 1b) (2, 7). All members of this family have a conserved helix-turn-helix (HTH) DNA-binding motif, with the same overall dimeric “winged” helix structure. Phylogenetic analysis suggests that this family evolved from a common evolutionary ancestor to sense specific types of metals (2, 17). However, the mechanisms of metal-induced DNA release do not seem to be the same. Furthermore, substantial variability exists in the metal-binding sites, which roughly can be divided up into two distinct types based on their location and the nature of the amino acid ligands. A few family members have both metal sites, although in these cases, the metal bound to one of the sites may act as a structural cofactor, and at least one other member possesses a divergent third site (18).
Several proteins in this family respond to metal binding in a cysteine-rich site with at least some of the ligands from the N-terminal helix (α3) that is a part of the HTH DNA-binding motif. This site, often referred to as α3, α3N, or site 1, controls a response to thiophilic ions such as cadmium, lead, or arsenite. For example, the arsenite bound by ArsR is coordinated by 2-3 cysteines (cys32, cys34, cys37) from the same α3 helix, so clustering the cysteines around arsenite would cause a large distortion of the helix that is proposed to disrupt DNA binding (19).
SmtB is regulated by the second type of metal-binding site called α5, α5C, or site 2. This site, which bridges the interface between α5 helices of the dimer, is composed of carboxylate and imidazole ligands and regulates the response to harder metal ions such as Co(II), Ni(II), and Zn(II). The structures of SmtB revealed that when both α5 sites of the dimer are filled with zinc, a significant change occurs in the 3° structure of the protein that compacts the molecule (Fig. 2a) (20). This zinc-dependent conformational switch likely is controlled by a hydrogen-bond network that links one metal-binding histidine to residues in the DNA-binding domain, which is a connection that is not apparent in the apo structure. It has been proposed that in the compact structure, the DNA-binding domains are too close together to bind properly to the spacing of the DNA recognition sequence, which results in release of the DNA and derepression.
A recent structure of apo-CadC suggests a different allosteric mechanism (21). CadC has both metal-binding sites, but the second site is probably structural because mutagenesis of the ligands in site 2 does not affect the metal-dependent response. The inducer site is α3N, and only two of the ligands (cys58 and cys60) are in the α3 HTH helix; the other two (cys7’ and cysll’) are donated by the N-terminus of the opposing subunit. Although the crystal structure does not have metal bound to the α3N site, the structure suggests that binding of the metal would pull the N-terminal strand toward the DNA-binding motif and block access to the DNA sterically block. However, mutants of the α3N ligands do not release the DNA even though they still bind metals tightly, albeit in altered coordination sites, which indicates that the allosteric response must be more complicated (22). Structures with metals bound to the regulatory site or in a complex with DNA will help to clarify the details of this system.
Figure 2. (a) Metal binding to SmtB. (Top) The holo-SmtB structure (dark gray, pdb 1R22) is significantly more compact than the apoprotein (light gray, pdb 1R1T). The zinc ions (spheres) are only observed in the α5 sites because the ligands of the α3 sites were mutated to generate a protein that is still functional in vivo but only binds one zinc ion per monomer. The putative DNA-binding helix of the HTH motif is indicated. (Bottom) The hydrogen-bonding network between one of the α5 zinc ions and the DNA-binding helix (L83) is highlighted. This network is not observed in the structure of the apoprotein. For clarity, the holo-protein is rotated slightly from the above view, and the two monomers are colored with different shades of gray. Putative hydrogens are indicated by thin bonds. (b) Structure of the E. coli nickel-responsive repressor NikR. The apoprotein (top, pdb 1Q5V), holo-protein (middle, pdb 2HZA), and DNA complex (bottom, pdb 2HCV) are shown with each monomer in the tetramer drawn in a different shade of gray. Two ribbon-helix-helix DNA-binding dimers flank the central core of four metal-binding domains. The nickel ions (smaller spheres) are coordinated in a square planar site by H87, H89, C95, and H76; of the opposing monomer (shown in inset, site rotated for clarity). The larger spheres in the DNA complex are best modeled as potassium ions. The sections of the metal-binding domain that contact the DNA are circled, and these regions are not well ordered in the apo-structure and could only be modeled in one of the four monomers.
NikR is the only known metalloregulator with a ribbon-helix-helix DNA-binding motif (5). This protein is a nickel- responsive repressor (Fig. 1a) that functions as a tetramer (Fig. 2b). It binds four nickel ions in square-planar sites that bridge the protein subunits, with the two flanking DNA-binding dimers linked to the central core of four metal-binding domains by flexible linkers. Several structures of E. coli NikR (apo-, holo-, and holo-DNA complexes) suggest an unusual mechanism for how nickel induces DNA binding (23). In the apo structure, the ribbons that contact the DNA are too far apart to bind the two sites in the palindromic recognition sequence concurrently. However, a comparison between the apo and holo structures revealed that nickel does not induce a rearrangement of the DNA-binding domains into an optimal conformation to interact with the DNA. Instead, nickel binding anneals a helix and a loop in the metal-binding domain that contact the DNA backbone, which suggests that the nickel stabilizes nonspecific protein-DNA interactions that facilitate binding of the DNA-binding domains the recognition sequences. A similar stabilization of secondary structure was observed in crystal structures of NikR from Pyrococcus horikoshii (24) and predicted by solution protease digestion experiments with the E. coli protein (25). Although biochemical evidence exists that additional nickel-binding sites strengthen DNA binding, it is yet not clear how this occurs.
The homologous iron-dependent regulators DtxR and IdeR control the expression of virulence factors as well as proteins involved in iron homeostasis in pathogenic and nonpathogenic bacteria (3, 26, 27). The DNA-binding domains are N-terminal HTHs, and the dimeric iron complexes bind to palindromic sequences in the promoters of regulated genes and repress transcription (Fig. 1a and Fig. 3). Two distinct metal-binding sites exist in each monomer. One site is referred to as ancillary because metal-dependent activity is much less sensitive to mutations at this site than of the primary-site ligands (See Reference 3 and references therein), although this site does have a role in enhancing the metal sensitivity of the repressor (28). Biochemical studies support a multistep, metal-activated DNA-binding mechanism that includes dimerization (29-32). Multiple X-ray crystal structures of the proteins without metals, in complexes with a variety of divalent metals bound in one or both sites, or bound to DNA, revealed that two dimers bind independently to opposite sides of the DNA (Fig. 3). Also, they suggest how the metal could influence DNA binding in addition to stabilizing the active dimer (for example, see References 33-36). The corepressor causes a shift in the DNA-binding domains in relation to the metal-binding domains, which closes the distance between these motifs and rotates them with respect to the rest of the molecule. One link between these two domains is Met10, which is a ligand of the primary-site metal from an N-terminal helix that contacts the DNA-recognition helix. The conformational change that allows Met10 to serve as a ligand also promotes hydrogen bonding between Glu9 and the imidazole of His106, another metal ligand. A hydrogen bond between His79 and Glu105, ligands of the ancillary and primary sites, respectively, support communication between the two metals. Furthermore, in the DtxR-DNA complexes, it was observed that the N-terminal helix is unwound so that it can be moved out of the way and avoid a steric clash with the DNA (34). This change in secondary structure may be mediated through hydrogen bonding with a metal-bound water.
Figure 3. Structure of the Co(II)-IdeR-DNA complex. (Top) Two IdeR dimers bind to the DNA recognition sequence, and each monomer (gray) binds two cobalt ions (spheres) (pdb 1U8 R). (Bottom) The two metal-binding sites in one monomer are connected by a hydrogen bond (not shown) between H79 and E105.
Organisms employ a variety of transition metals, each for distinct functions. Although indirect connections exist between the metal pathways, it seems that the direct cellular control of individual metals, both nutrients and environmental toxins, functions independently. Each metalloregulator is dedicated to one metal or a subset of metal ions, even though it may belong to the same family as other metalloregulators. It is becoming clear that to differentiate between the available metals these proteins can take advantage of the inorganic chemistry of the metals ions by a variety of means (18).
Metal selection by protein sites
The high-resolution structures of the metal-binding sites of two MerR homologs, CueR and ZntR, reveal very distinctive coordination environments, in part controlled by the number of cysteine residues (Fig. 4) (37). CueR responds to Cu(I), Ag(I), and Au(I), and it binds all three metals in a linear, two-coordinate site composed of two cysteine residues that are conserved in all members of this family. This low coordination number is preferred by metals in the +1 oxidation state, and the site is shielded to prevent expansion of the coordination number with external ligands. In addition, the authors suggest that CueR can discriminate against Hg(II), which is one of the few divalent metals that binds favorably to linear dithiolate sites, because it is optimized to provide charge compensation for a +1 metal ion but not +2. ZntR, which responds to Zn(II), Cd(II), and Pb(II), has the same overall dimeric structure as CueR, but it has four cysteines in each metal-binding site that contains a dinuclear zinc cluster. Additional ligands include a histidine and a phosphate ion (or sulfate). One of these cysteines, Cys79, is from the opposite end of the dimerization helix of the opposing subunit, which is linked to the DNA-binding domain of that monomer and suggests a mechanism of communication between the two domains. However, this cysteine is conserved only in the homologs that respond to divalent metals, which prefer higher coordination numbers and require larger charge neutralization than the monovalent metals.
Although the structure of Hg(II)-MerR has not been reported yet, a sequence alignment indicates that only three of the four cysteine ligands of ZntR are conserved in MerR. Furthermore, spectroscopic and mutagenesis analysis demonstrated that the metal is bound to these three cysteines (7). This intermediate number of cysteines suggests how MerR can select against the monovalent ions, which prefer the dithiolate sites and would not provide as much charge neutralization, as well as the divalent metals such as Zn(II) that prefer higher coordination numbers.
Figure 4. (Top) Structure of CueR (pdb 1Q05) with two copper ions bound. One metal site is blown up in the picture below and has been rotated slightly for clarity. Ser77’ is on the opposite subunit from the copper ligands and is replaced by a cysteine in ZntR. (Bottom) Metal-binding site of the dizinc cluster in ZntR (pdb 1Q08). The two subunits are drawn in light and dark gray ribbons.
Metal selectivity through the metals
Given that the coordination geometry of a metal complex is a determinant of metal selectivity, it is predicted that all metalloregulators would have a metal-binding site preorganized to accept the appropriate metal(s) while excluding all others. However, this clearly is not always the case because many examples of metalloregulators exist in which the metal ligands could accommodate multiple different metals. Furthermore, if this model were correct, the affinity of the different metals would parallel the selectivity for the DNA-binding response, which is not always observed. Instead, in some cases, the metal-binding sites are clearly flexible and the presence of the correct metal is recognized by the distinct coordination geometry imposed on the protein.
For example, NmtR from Mycobacterium tuberculosis is a member of the ArsR/SmtB family that responds in vivo to nickel, and to a lesser extent to cobalt, which binds in an α5 site. Zinc is a poor allosteric inducer both in vitro and in vivo, even though it binds more tightly to the protein than nickel and cobalt (38). An explanation for these observations was provided by spectroscopic analysis that revealed the Zn(II) ion bound in a tetrahedral 4-coordinate site, whereas Ni(II) was bound in a 6-coordinate octahedral site (38, 39). In contrast, CzrA from Staphylococcus aureus responds well to zinc, not nickel, but this protein binds the different metal ions in an a5 site with the same type of geometries as in NmtR (39). Thus, these two proteins accommodate each metal ion in the preferred coordination geometries of the metals, but they have evolved such that only one activates the allosteric response of each protein: 4-coordinate activates CzrA and 6-coordinate activates NmtR.
One question that is raised by these studies is whether a metal ion that binds tightly to the protein but is a poor allosteric effector in vitro will compete with the inducer and act as an inhibitor. This issue is currently under investigation (40), and one factor that clearly must be addressed is whether all of the possible metals are even available in vivo.
When NmtR from M. tuberculosis was transplanted to a cyanobacterial host, a response from the protein was observed when extra cobalt was added to the growth media but not to nickel (38). Metal analysis revealed that both metals were imported into the native M. tuberculosis cells, but only cobalt was taken up substantially by the cyanobacteria, which explains the lack of response to nickel in the heterologous host. Similarly, mutations in DtxR designed to alter the selectivity from iron to manganese resulted in a decrease in iron responsiveness in vivo that was restored partially by increasing iron uptake and availability (41).
The flip side of this issue is the level of sensitivity of the metalloregulators; i.e., how much metal is necessary to activate the genetic switch? For example, an in vitro study of a pair of E. coli zinc sensors, ZntR and Zur, revealed a sensitivity that correlated with physiologic function (42). In response to zinc, Zur shuts off transcription of zinc uptake genes and ZntR turns on transcription of efflux. They have a graded response such that Zur responds to lower zinc concentrations than ZntR, turning off uptake before efflux is activated, which prevents both uptake and efflux from working against each other concurrently in a futile cycle.
It is also interesting to note that the responses of ZntR and Zur both occur over a very small gradient of zinc concentrations (≈2 orders of magnitude), which reveals the tight window of optimal zinc levels that falls between starvation and toxicity. However, the metal-buffered conditions suggest that this optimum is on the order of femtomolar free zinc, which is far less than one zinc ion per cell. CueR, which activates Cu(I) export, may be even more sensitive (37). Essentially, such sensitivities imply that no free copper or zinc exists in an E. coli cell under healthy growth conditions and invite questions about the availability of the essential metals for the destination biomolecules that use the metals as cofactors, many of which have much weaker thermodynamic affinities for the metals than the metalloregulators. It has been proposed that in a cellular context, metal delivery is under kinetic control (43), possibly through the activity of intracellular trafficking factors. In addition, many factors exist in a cellular milieu with varying degrees of metal-binding capabilities, such as amino acids, carbohydrate metabolites, thiol-containing molecules, and even weak nonspecific sites on protein surfaces, which would soak up any “free” metal ions and would provide pools of readily accessible ions if needed. The metalloregulators must be tuned to this buffering capacity to respond appropriately and to maintain a healthy balance of metal homeostasis.
Tools and Techniques
To define the complete mechanisms of the metalloregulators requires the use of a broad spectrum of methods, some of which are mentioned below. The first clue that a gene or an operon is controlled by a metalloregulator can come from in vivo experiments that demonstrate a change in expression when the organism is grown in the presence of extra metal. Then, the specific metalloregulators can be found through genetic experiments. Another approach, facilitated by the many complete genome sequences now available, is to search for a gene that encodes a homolog of a known metalloregulator, or to examine operons that encode proteins that are clearly involved in metal homeostasis. Once a gene is identified as a possible metalloregulator, the assignment must be confirmed by in vitro experiments.
After recombinant expression and purification of the protein, several different methods are used to examine metal binding and how this affects DNA (or RNA) binding. The protein might be purified with some metal bound, but whether that metal is physiologically relevant or one that was available in the expression host is a tricky question and should be resolved by in vivo experiments. Some metal-protein sites can be observed by electronic absorption spectroscopy (UV/visible spectroscopy), which can be used to determine the stoichiometry if the affinity is tight enough, as well as the dissociation constant, possibly through the use of small-molecule chelators as competitors. Metal binding and stoichiometry can also be examined by treating the protein with excess metal, removal of unbound metal with a method such as dialysis or gel filtration chromatography, and then direct metal analysis with inductively coupled plasma atomic emission spectroscopy ICP-AES, ICP-MS, atomic absorption spectroscopy, or other techniques. Many standard procedures to measure the strength of a protein-ligand interaction can be used to determine the metal affinity; in addition to UV/visible spectroscopy, these procedures include fluorescence spectroscopy, isothermal titration calorimetry, and equilibrium dialysis.
Conserved residues, particularly cysteine, histidine, and aspartate/glutamate, can signal likely metal ligands, which is an assignment that can be tested by mutagenesis. Detailed information on the coordination sphere can also be provided by spectroscopic techniques such as X-ray absorption spectroscopy, as well as UV/visible spectroscopy, or electronic paramagnetic resonance spectroscopy for some metals.
To examine DNA binding, typically a pair of complementary oligonucleotides that contain the DNA recognition sequence is used. If the binding site is not known, a method such as DNAse footprinting on a longer fragment of DNA that contains the whole promoter region will reveal the location of the binding site. The recognition sequence can be confirmed by using in vivo reporter assays. DNAse footprinting, mobility shift assays, and fluorescence anisotropy are some common techniques used to monitor DNA binding in the presence or absence of metal(s). Variations on these methods can provide information on whether the protein bends or unwinds the DNA. In the case of metalloregulators that function as transcription factors, in vitro transcription assays can sometimes be used to examine directly how the protein and metal influence transcription.
Finally, biophysical studies can examine how metal binding influences protein conformation and can serve as the basis for a hypothesis about the connection between the two activities, metal binding and DNA binding. Furthermore, it is clear from the discussion that a high-resolution structure, either from X-ray crystallography or nuclear magnetic resonance spectroscopy, is indispensable. A structure of the apo protein provides information on likely metal sites and the conformation of the DNA-binding domain. Structures of the metal-bound and/or DNA-bound complexes, in comparison with the isolated protein, can illuminate the molecular details of the structure-function relationship and can serve as a key reference point in understanding all of the pieces of information provided by solution studies.
Although the mechanisms of some types of metalloregulators are becoming clear, many questions about other systems remain. In addition, we are now at the point at which we can start to address more general issues. For example, how does the metal affinity of the isolated metalloregulator fit into the context of the competing surroundings of the cell? Do metal chaperones that deliver the metal to the regulators exist, as in the case of the cop system? How selective must the regulator be given the limited availability of certain metals? Do specific factors exist that reverse the effects of metal binding to the metalloregulators, a role suggested for MerD, or are the routine protein degradation and production pathways sufficient? What is the connection and means of communication between the different metal pathways or between a given metal pathway and other cellular systems? Answers to these questions will expand our understanding of the role of metalloregulators in the complex and dynamic cellular environment.
1. O’Halloran TV. Transition metals in control of gene expression. Science 1993; 261:715-725.
2. Busenlehner LS, Pennella MA, Giedroc DP. The SmtB/ArsR family of metalloregulatory transcriptional repressors: structural insights into prokaryotic metal resistance. FEMS Microbiol. Rev. 2003; 27:131-143.
3. Rodriguez GM, Smith I. Mechanisms of iron regulation in mycobacteria: role in physiology and virulence. Mol. Microbiol. 2003; 47:1485-1494.
4. Andrews SC, Robinson AK, Rodriguez-Quinones F. Bacterial iron homeostasis. FEMS Microbiol. Rev. 2003; 27:215-237.
5. Dosanjh NS, Michel SLJ. Microbial nickel metalloregulation: NikRs for nickel ions. Curr. Opin. Chem. Biol. 2006; 10:1-8.
6. Outten FW, Outten CE, O’Halloran TV. Metalloregulatory systems at the interface between bacterial metal homeostasis and resistance. In: Bacterial Stress Responses. Storz G, Hengge-Aronis R, eds. 2000. ASM Press, Washington D.C. pp. 145-157.
7. Brown NJ, Stoyanov JV, Kidd SP, Hobman JL. The MerR family of transcriptional regulators. FEMS Microbiol. Rev. 2003; 27:145- 163.
8. Champier L, Duarte V, Michaud-Soret I, Coves J. Characterization of the MerD protein from Ralstonia metallidurans CH34: A possible role in bacterial mercury resistance by switching off the induction of the mer operon. Mol. Microbiol. 2004; 52:1475-1485.
9. Hentze MW, Kuhn LC. Molecular control of vertebrate iron metabolism: mRNA-based regulatory circuits operated by iron, nitric oxide, and oxidative stress. Proc. Natl. Acad. Sci. U.S.A. 1996; 93:8175-8182.
10. Thomson AM, Rogers JT, Leedman PJ. Iron-regulatory proteins, iron-responsive elements and ferritin mRNA translation. Int. J. Biochem. Cell. Biol. 1999; 31:1139-1152.
11. Hentze MW, Muckenthaler MU, Andrews NC. Balancing acts: Molecular control of mammalian iron metabolism. Cell 2004; 117:285-297.
12. Hintze KJ, Theil EC. Cellular regulation and molecular interactions of the ferritins. Cell. Mol. Life Sci. 2006; 63:591-600.
13. Masse E, Arguin M. Ironing out the problem: new mechanisms of iron homeostasis. Trends Biochem. Sci. 2005; 30:462-468.
14. Morita T, Mochizuki Y, Aiba H. Translational repression is sufficient for gene silencing by bacterial small noncoding RNAs in the absence of mRNA destruction. Proc. Natl. Acad. Sci. U.S.A. 2006; 103:4858-4863.
15. Rensing C, Grass G. Escherichia coli mechanisms of copper homeostasis in a changing environment. FEMS Microbiol. Rev. 2003; 27:197-213.
16. Solioz M, Stoyanov JV. Copper homeostasis in Enterococcus hirae. FEMS Microbiol. Rev. 2003; 27:183-195.
17. Tottey S, Harvie DR, Robinson NJ. Understanding how cells allocate metals using metal sensors and metallochaperones. Acc. Chem. Res. 2005; 38:775-783.
18. Pennella MA, Giedroc DP. Structural determinants of metal selectivity in prokaryotic metal-responsive transcriptional regulators. BioMetals 2005; 18:413-428.
19. Xu C, Rosen BP. Metalloregulation of soft metal resistance pumps. In: Metals and Genetics. Sarkar B, ed. 1999. Kluwer Academic/Plenum publishers, New York. pp. 5-19.
20. Eicken C, Pennella MA, Chen X, Koshlap KM, Van Zile ML, Sacchettini JC, Giedroc DP. A metal-ligand-mediated intersubunit allosteric switch in related SmtB/ArsR zinc sensor proteins. J. Mol. Biol. 2003; 333:683-695.
21. Ye J, Kandegedara A, Martin P, Rosen BP. Crystal structure of the Staphylococcus aureus pI258 CadC Cd(II)/Pb(II)/Zn(n)- responsive repressor. J. Bacteriol. 2005; 187:4214-4221.
22. Busenlehner LS, Weng T-C, Penner-Hahn JE, Giedroc DP. Elucidation of primary (α3N) and vestigial (α5) heavy metal-binding sites in Staphylococcus aureus pI258 CadC: evolutionary implications for metal ion selectivity of ArsR/SmtB metal sensor proteins. J. Mol. Biol. 2002;319:685-701.
23. Schreiter ER, Wang SC, Zamble DB, Drennan CL. NikR-operator complex structure and the mechanism of repressor activation by metal ions. Proc. Natl. Acad. Sci. U.S.A. 2006; 103:13676-13681.
24. Chivers PT, Tahirov TH. Structure of Pyrococcus horikoshii NikR: nickel sensing and implications for the regulation of DNA recognition. J. Mol. Biol. 2005; 348:597-607.
25. Dias AV, Zamble DB. Protease digestion analysis of Escherichia coli NikR: Evidence for conformational stabilization with Ni(II). J. Biol. Inorg. Chem. 2005; 10:605-612.
26. Holmes RK. Biology and molecular epidemiology of diphtheria tox and the tox gene. J. Infect. Diseases 2000; 181:S156-S167.
27. Ranjan S, Yellaboina S, Ranjan A. IdeR in mycobacteria: from target recognition to physiological function. Crit. Rev. Microbiol. 2006; 32:69-75.
28. Love JF, vander Spek JC, Marin V, Guerrero L, Logan TM, Murphy JR. Genetic and biophysical studies of diptheria toxin repressor (DtxR) and the hyperactive mutant DtxR(E175K) support a multistep model of activation. Proc. Natl. Acad. Sci. U.S.A. 2004; 101:2506-2511.
29. Spiering MM, Ringe D, Murphy JR, Marietta MA. Metal stoichiometry and functional studies of the diphtheria toxin repressor. Proc. Natl. Acad. Sci. U.S.A. 2003; 100:3808-3813.
30. Rangachari V, Marin V, Bienkiewicz EA, Semavina M, Guerrero L, Love JF, Murphy JR, Logan TM. Sequence of ligand binding and structure change in the diphtheria toxin repressor upon activation by divalent transition metals. Biochemistry 2005; 44:5672-5682.
31. D’Aquino JA, Tetenbaum-Novatt J, White A, Berkovitch F, Ringe D. Mechanism of metal ion activation of the diphtheria toxin repressor DtxR. Proc. Natl. Acad. Sci. U.S.A. 2005; 102:18408- 18413.
32. Semavina M, Beckett D, Logan TM. Metal-linked dimerization in the iron-dependent regulator from Mycobacterium tuberculosis. Biochemistry 2006; 45:12480-12490.
33. Pohl E, Holmes RK, Hol WGJ. Motion of the DNA-binding domain with respect to the core of the diphtheria toxin repressor (DtxR) revealed in the crystal structures of apo- and holo-DtxR. J. Biol. Chem. 1998; 273:22420-22427.
34. White A, Ding X, vander Spek JC, Murphy JR, Ringe D. Structure of the metal-ion-activated diphtheria toxin repressor/tox operator complex. Nature 1998; 394:502-506.
35. Pohl E, Holmes RK, Hol WGJ. Crystal structure of the iron- dependent regulator (IdeR) from Mycobacterium tuberculosis shows both metal binding sites fully occupied. J. Mol. Biol. 1999; 285:1145-1156.
36. Wisedchaisri G, Holmes RK, Hol WGJ. Crystal structure of an IdeR-DNA complex reveals a conformational change in activated IdeR for base-specific interactions. J. Mol. Biol. 2004; 342:1155- 1169.
37. Changela A, Chen K, Xue Y, Holschen J, Outten CE, O’Halloran TV, Mondragon A. Molecular basis of metal-ion selectivity and zeptomolar sensitivity by CueR. Science 2003; 301:1383-1387.
38. Cavet JS, Meng W, Pennella MA, Appelhoff RJ, Giedroc DP, Robinson NJ. A nickel-cobalt-sensing ArsR-SmtB family repressor. J. Biol. Chem. 2002; 277:38441-38448.
39. Pennella MA, Shokes JE, Cosper NJ, Scott RA, Giedroc DP. Structural elements of metal selectivity in metal sensor proteins. Proc. Natl. Acad. Sci. U.S.A. 2003; 100:3713-3718.
40. Harvie DR, Andreini C, Connolly BA, Yoshida K, Fujita Y, Harwood CR, Radford DS, Tottey S, Cavet JS, Robinson NJ. Predicting metals sensed by ArsR-SmtB repressors: Allosteric interference by a non-effector metal. Mol. Microbiol. 2006; 59:1341-1356.
41. Guedon E, Helmann JD. Origins of metal ion selectivity in the DtxR/MntR family of metalloregulators. Mol. Microbiol. 2003; 48:495-506.
42. Outten CE, O’Halloran TV. Femtomolar sensitivity of metalloregulatory proteins controlling zinc homeostasis. Science 2001; 292:2488-2492.
43. Finney LA, O’Halloran TV. Transition metal speciation in the cell: insights from the chemistry of metal ion receptors. Science 2003; 300:931-936.
Andrews GK. Cellular zinc sensors: MTF-1 regulation of gene expression. BioMetals 2001; 14:223-237.
Balamurugan K, Schaffner W. Copper homeostasis in eukaryotes: Teetering on a tightrope. Biochim. Biophys. Acta 2006; 1763:737-746.
Cavet JS, Borrelly GPM, Robinson NJ. Zn, Cu and Co in cyanobacteria: selective control of metal availability. FEMS Microbiol. Rev. 2003; 27:165-181.
Rutherford JC, Bird AJ. Metal-responsive transcription factors that regulate iron, zinc, and copper homeostasis in eukaryotic cells. Eukaryotic Cell 2004; 3:1-13.
Sarkar B, ed. Metals in Genetics. 1999. Kluwer Academic/Plenum Publishers, New York.
Silver S, Walden W, eds. Metal Ions in Gene Regulation. 1998. International Thomson Publishing, New York.
Winge DR, Jensen LT, Srinivasan C. Metal-ion regulation of gene expression in yeast. Curr. Opin. Chem. Biol. 1998; 2:216-221.
Metal Complexes, Assembly of
Metal Homeostasis, Intercellular
Metal Homeostasis: An Overview
Metal Transport Through Membranes
Metallochaperones, Chemistry of