The protein tristetraprolin (TTP, also known as NUP475 and TIS11) is a nonclassical zinc finger protein that is involved in regulating the inflammatory response. Specifically, TTP binds to AU-rich sequence elements located at the 3'-untranslated region of cytokine mRNAs forming a complex that is degraded by the exosome. The nucleic acid binding region of TTP is comprised of two CysX(8)CysX(5)CysX(3)His domains that are activated in the presence of zinc. A two-domain construct of TTP (TTP-2D) has been cloned and overexpressed in E. coli. TTP-2D picks up visible red coloration from the expression media, unless it is expressed under iron-restricted conditions. The iron-binding properties of TTP-2D and the effect of iron substitution on RNA recognition have been investigated. Both Fe(II) and Fe(III) bind to TTP-2D and a full titration of Fe(III) with TTP-2D revealed that this metal ion binds with micromolar affinity. Upon reconstitution of TTP-2D with either Fe(II) or Fe(III), the protein recognizes a canonical RNA-binding sequence, UUUAUUUAUUU, with nanomolar affinity. Substitution of a single adenine or both adenines results in a decreased affinity of TTP-2D for the RNA molecule, demonstrating that both Fe(II)-TTP-2D and Fe(III)-TTP-2D selectively recognize a physiologically relevant RNA sequence. The relative affinities of Fe(II)-TTP-2D and Fe(III)-TTP-2D for the series of RNA sequences mirror those observed for Zn(II)-TTP-2D and suggest that iron is a viable substitute for zinc in this protein.

HPNikR is a prokaryotic nickel binding transcription factor found in the virulent bacterium Helicobacter pylori. HPNikR regulates the expression of multiple genes as an activator or repressor, including those involved in nickel ion homeostasis, acid adaptation, and iron uptake. The target operator sequences of the genes regulated by HPNikR do not contain identifiable symmetrical recognition sites, and the mechanism by which HPNikR distinguishes between the genes it regulates is not understood. Using competitive fluorescence anisotropy (FA) and electrophoretic gel mobility shift (EMSA) assays, the interactions between HPNikR and the target operator sequences of the genes directly regulated (ureA, NixA, NikR, Fur OPI, Fur OPII, Frpb4, FecA3, and exbB) were characterized. These studies revealed that HPNikR utilizes a two-tiered mode of DNA recognition by binding to some genes with high affinity and others with low affinity. The genes that are tightly regulated by HPNikR encode proteins that utilize nickel, while those that are less tightly regulated encode other types of proteins. The affinities of low-affinity metal ions for a second metal binding site were determined to be in the micromolar regime, and a contribution of electrostatics to the HPNikR-DNA binding event was determined. Detailed studies of the role of sequence length and identity for the interaction between HPNikR and ureA revealed a specific length requirement for DNA binding.

Zinc finger (ZF) proteins are a large family of metalloproteins that utilize zinc for structural purposes. Zinc coordinates to a combination of cysteine thiol and histidine imidazole residues within the ZF polypeptide sequence resulting in a folded and functional protein. Initially, a single class of ZFs were identified. These ZFs, now referred to as the "classical" ZFs, utilize a Cys2His2 (CCHH) ligand set to bind zinc. Upon Zn coordination, the classical ZFs fold into a structure made up of an α helix and an antiparallel β sheet. When folded, classical ZFs recognize and bind to specific DNA targets and function as transcription factors. With the advent of genome sequencing and proteomics, many additional classes of ZFs were identified based upon their primary amino acid sequences. At least 13 additional classes of ZFs are known, and collectively these "nonclassical" ZFs differ in the ligand set involved in Zn(II) coordination, the organization of the ligands within the polypeptide sequence and the macromolecular targets. Some nonclassical ZFs are DNA binding "transcription factors", while others are involved in RNA regulation and protein recognition. Much less is known about these nonclassical ZFs with regards to the roles of metal coordination in fold and function. This Account focuses on our laboratory's efforts to characterize two families of "nonclassical" ZFs: the Cys3His (or CCCH) ZF family and the Cys2His2Cys (or CCHHC) ZF family. Our work on the CCCH ZF family has focused on the protein Tristetraprolin (TTP), which is a key protein in regulating inflammation. TTP contains two CCCH domains that were proposed to be ZFs based upon their sequence. We have shown that while this protein can coordinate Zn(II) at the CCCH sites, it can also coordinate Fe(II) and Fe(III). Moreover, the zinc and iron bound forms of TTP are equally adept at discriminating between RNA targets, which we have demonstrated via a fluorescence anisotropy based approach. Thus, CCCH type ZFs appear to be promiscuous with respect to metal preference and a role for iron coordination in CCCH ZF function is proposed. The CCHHC family of ZFs is a small family of nonclassical ZFs that are essential for the development of the central nervous system. There are three ZFs in this family: neural zinc finger factor-1 (NZF-1), myelin transcription factor-1 (MyT1), and suppressor of tumorgenicity 18 (ST18). All three proteins contain multiple clusters of "CCHHC" domains, which are all predicted to be Zn binding domains. We have focused on a tandem-CCHHC domain construct of NZF-1, which recognizes β-RARE DNA, and we have identified key residues required for DNA recognition. Unlike classical ZFs, for which a few conserved residues are required for DNA recognition, the CCHHC class of ZFs utilize a few nonconserved residues to drive DNA recognition leading us to propose a new paradigm for ZF/DNA binding.

Zinc finger proteins utilize zinc for structural purposes: zinc binds to a combination of cysteine and histidine ligands in a tetrahedral coordination geometry facilitating protein folding and function. While much is known about the classical zinc finger proteins, which utilize a Cys(2)His(2) ligand set to coordinate zinc and fold into an anti-parallel beta sheet/alpha helical fold, there are thirteen other families of 'non-classical' zinc finger proteins for which relationships between metal coordination and protein structure/function are less defined. This 'Perspective' article focuses on two classes of these non-classical zinc finger proteins: Cys(3)His type zinc finger proteins and Cys(2)His(2)Cys type zinc finger proteins. These proteins bind zinc in a tetrahedral geometry, like the classical zinc finger proteins, yet they adopt completely different folds and target different oligonucleotides. Our current understanding of the relationships between ligand set, metal ion, fold and function for these non-classical zinc fingers is discussed.

HPNikR, a prokaryotic nickel binding transcription factor, is found in Helicobacter pylori where it functions as a regulator of multiple genes, including those involved in acid adaptation and nickel ion homeostasis. Particularly important is HPNikR's role in the regulation of the nickel-dependent enzyme urease which is critical for the organism's survival in the acidic environment of the gastric epithelium. The target operator sequences of the genes regulated by HPNikR do not contain any identifiable palindromes, and the exact mechanism(s) of the HPNikR-DNA recognition event is unknown. HPNikR was expressed and purified as a soluble protein containing mixed alpha/beta secondary structure with evidence of a tertiary fold. A direct and competitive fluorescence anisotropy (FA) assay to probe both the metal ion requirements and sequence specificity of HPNikR for PureA, the operator sequence for the urease gene, was developed. FA studies revealed that apo-HPNikR did not bind to PureA while Ni(II)HPNikR bound PureA with nanomolar affinity, but only in the presence of a second metal ion [magnesium, calcium, or manganese(II)], suggesting that HPNikR contains a second, low-affinity metal binding site. Cu(II)HPNikR also exhibited a requirement for a second metal ion to accomplish PureA binding. Removal of a loosely conserved "putative" palindrome sequence in the PureA operator abrogated HPNikR binding. Together, these results support a model of HPNikR-PureA binding in which specific metal ions must be coordinated to high- and low-affinity sites to modulate binding.

No abstract

Regulation of gene expression takes place at several different levels and involves specific domains involved in specific protein-nucleic acid interactions. The protein Nup475 (also known as Tristetraprolin and TS11) binds to AU-rich sequence elements in certain mRNA molecules and favors the degradation of these mRNAs. The nucleic acid binding domain of Nup475 consists of two CCCH zinc-binding domains. A 36-amino acid peptide corresponding to the first of these CCCH domains has been synthesized and characterized. This peptide binds metal ions such as zinc(II) and cobalt(II) with affinities comparable to those of other authenticated zinc-binding domains. The zinc(II) complex of this peptide binds the RNA oligonucleotide UUUAUUU labeled with fluorescein on the 3'-end with an affinity of approximately 5 microM and discriminates against other sequences lacking the central A or the flanking U residues. These results demonstrate for the first time that a single CCCH domain is capable of binding single-stranded RNA with considerable affinity and selectivity. The combination of this well-behaved domain and the fluorescence-based binding assay sets the stage for more detailed structure-activity studies.

No abstract