The past, present, and future of artificial zinc finger proteins: design strategies and chemical and biological applications.

Zinc finger proteins are abundant in the human proteome and are responsible for a variety of functions. The domains that constitute zinc finger proteins are compact spherical structures, each comprising approximately 30 amino acid residues, but they also have precise molecular factor functions: zinc binding and DNA recognition. Due to the biological importance of zinc finger proteins and their unique structural and functional properties, many artificial zinc finger proteins have been created and are expected to improve their functions and biological applications. In this study, we review previous studies on the redesign and application of artificial zinc finger proteins, focusing on the experimental results obtained by our research group. In addition, we systematically review various design strategies used to construct artificial zinc finger proteins and discuss in detail their potential biological applications, including gene editing. This review will provide relevant information to researchers involved or interested in the field of artificial zinc finger proteins as a potential new treatment for various diseases.

Importance of two-dimensional cation clusters induced by protein folding in intrinsic intracellular membrane permeability.

We investigated the cell penetration of Sp1 zinc finger proteins (Sp1 ZF) and the mechanism via which the total cationic charge and distribution of cationic residues on the protein surface affect intracellular trafficking. Sp1 ZFs showed intrinsic cell membrane permeability. The intracellular transfer of Sp1 ZFs other than 1F3 was dependent on the total cationic charge. Investigation of the effect of cationic residue distribution on intracellular membrane permeability revealed that the cellular uptake of unfolded Zn2+-non-coordinating Ala mutants was lower than that of the wild type. Therefore, the total cationic charge and distribution of cationic residues on the protein played crucial roles in intracellular translocation. Mutational studies revealed that the two-dimensional cation cluster on the protein surface significantly improved their cellular uptake. This study will contribute to the design of artificial cargoes that can efficiently transport target substances into cells.

Feedback Response to Selective Depletion of Endogenous Carbon Monoxide in the Blood.

The physiological roles of endogenous carbon monoxide (CO) have not been fully understood because of the difficulty in preparing a loss-of-function phenotype of this molecule. Here, we have utilized in vivo CO receptors, hemoCDs, which are the supramolecular 1:1 inclusion complexes of meso-tetrakis(4-sulfonatophenyl)porphinatoiron(II) with per-O-methylated ß-cyclodextrin dimers. Three types of hemoCDs (hemoCD1, hemoCD2, and hemoCD3) that exhibit different CO-affinities have been tested as CO-depleting agents in vivo. Intraperitoneally administered hemoCD bound endogenous CO within the murine circulation, and was excreted in the urine along with CO in an affinity-dependent manner. The sufficient administration of hemoCD that has higher CO-affinity than hemoglobin (Hb) produced a pseudoknockdown state of CO in the mouse in which heme oxygenase-1 (HO-1) was markedly induced in the liver, causing the acceleration of endogenous CO production to maintain constant CO-Hb levels in the blood. The contents of free hemin and bilirubin in the blood plasma of the treated mice significantly increased upon removal of endogenous CO by hemoCD. Thus, a homeostatic feedback model for the CO/HO-1 system was proposed as follows: HemoCD primarily removes CO from cell-free CO-Hb. The resulting oxy-Hb is quickly oxidized to met-Hb by oxidant(s) such as hydrogen peroxide in the blood plasma. The met-Hb readily releases free hemin that directly induces HO-1 in the liver, which metabolizes the hemin into iron, biliverdin, and CO. The newly produced CO binds to ferrous Hb to form CO-Hb as an oxidation-resistant state. Overall, the present system revealed the regulatory role of CO for maintaining the ferrous/ferric balance of Hb in the blood.

Efficient cleavage of DNA oligonucleotides by a non-FokI-type zinc finger nuclease containing one His4-type finger domain derived from the first finger domain of Sp1.

In this study, we sought to improve the hydrolytic activity of a His4-type single finger domain (f2), which was previously derived from the second finger domain (f2') of the Sp1 zinc finger protein (Sp1wt), which has 3 tandem finger domains (f1', f2', and f3'). To this end, 2 His4-type single finger domains were generated by mutating 2 Cys residues participating in Zn(II) coordination with the His residues in the first (f1') and third finger (f3') domains of Sp1wt. Circular dichroism spectroscopy results showed that the first and second His4-type zinc finger domains (f1 and f2) adopted folded ßßα structures in the presence of Zn(II), but that the third His4-type zinc finger domain (f3) did not. Non-FokI-type zinc finger nucleases containing 3 or 4 finger domains were also prepared by combining a His4-type zinc finger domain with the Sp1wt scaffold. We studied their DNA-binding abilities and hydrolytic activities against DNA oligonucleotides by performing gel-mobility-shift assays. The results showed that f1 had higher hydrolytic activity for a DNA oligonucleotide with a GC box (5'-GGG GCG GGG-3'), compared with that of f2, although both His4-type single finger domains had similar DNA-binding affinities. The difference in the hydrolytic activity between f1 and f2 was ascribed not only to the zinc coordinate structure, but also to its folding structure and the stability of finger domain.

Intrinsic cell permeability of the GAGA zinc finger protein into HeLa cells.

We examined the intrinsic cell permeability of a GAGA zinc finger obtained from the Drosophila melanogaster transcription factor and analyzed its mechanism of cellular uptake using confocal microscopy and flow cytometry. HeLa cells were treated with the Cy5-labeld GAGA peptides (containing a fluorescent chromophore) to detect fluorescence signals from the fluorescent labeling peptides by confocal microscopy. The results clearly indicated that GAGA peptides possess intrinsic cell permeability for HeLa cells. Based on the results of the flow cytometry analysis and the theoretical net positive charge of the GAGA peptides, the efficiency of cellular uptake of the GAGA peptides was predicted to depend on the net positive charge of the GAGA peptide as well as the cationic component ratio of Arg residues to Lys residues.

Supramolecular intracellular delivery of an anionic porphyrin by octaarginine-conjugated per-O-methyl-ß-cyclodextrin.

A convenient and efficient method for intracellular delivery of a water-soluble anionic porphyrin has been developed by utilizing its supramolecular interaction with per-O-methyl-ß-cyclodextrin bearing an octaarginine chain as a cell-penetrating peptide.

Zn(II) binding and DNA binding properties of ligand-substituted CXHH-type zinc finger proteins.

CCHH-type zinc fingers are among the most common DNA binding motifs found in eukaryotes. In a previous report, we substituted the second ligand cysteine residue with aspartic acid, producing a Zn(II)-responsive transcription factor; this indicates that a ligand substitution is a possible design target of an engineered zinc finger peptide. Despite the importance of Zn(II) binding with respect to the folding and DNA binding properties of a zinc finger peptide, no study about the effects of ligand substitution on both Zn(II) binding and DNA binding properties has been reported. Here, we substituted a conserved cysteine (C) with other zinc-coordinated amino acid residues, histidine (H), aspartic acid (D), and glutamic acid (E), to create CXHH-type zinc finger peptides (X = C, H, D, and E). The Zn(II)-dependent conformational change was observed in all peptides; however, the Zn(II) binding affinity and metal coordination geometry of the peptides were different. Gel mobility shift assays showed that the Zn(II)-bound forms of the ligand-substituted derivatives retain DNA binding ability, while the DNA binding affinity decreased in the following manner: CCHH > CDHH > CEHH ≫ CHHH. The DNA binding sequence preferences of the ligand-substituted derivatives were similar to that of the wild type in the context of the full three-finger DNA-binding domain of transcription factor Zif268. These results indicate that artificial zinc finger proteins with various DNA binding affinities that respond to a diverse range of Zn(II) concentrations can be designed by substituting the Zn(II) ligand.

Rational design of DNA sequence-specific zinc fingers.

We developed a rational scheme for designing DNA binding proteins. The scheme was applied for a zinc finger protein and the designed sequences were experimentally characterized with high DNA sequence specificity. Starting with the backbone of a known finger structure, we initially calculated amino acid sequences compatible with the expected structure and the secondary structures of the designed fingers were then experimentally confirmed. The DNA-binding function was added to the designed finger by reconsidering a section of the amino acid sequence and computationally selecting amino acids to have the lowest protein-DNA interaction energy for the target DNA sequences. Among the designed proteins, one had a gap between the lowest and second lowest protein-DNA interaction energies that was sufficient to give DNA sequence-specificity.

An arginine residue instead of a conserved leucine residue in the recognition helix of the finger 3 of Zif268 stabilizes the domain structure and mediates DNA binding.

The Cys(2)His(2)-type zinc finger is a common DNA binding motif that is widely used in the design of artificial zinc finger proteins. In almost all Cys(2)His(2)-type zinc fingers, position 4 of the α-helical DNA-recognition site is occupied by a Leu residue involved in formation of the minimal hydrophobic core. However, the third zinc finger domain of native Zif268 contains an Arg residue instead of the conserved Leu. Our aim in the present study was to clarify the role of this Arg in the formation of a stable domain structure and in DNA binding by substituting it with a Lys, Leu, or Hgn, which have different terminal side-chain structures. Assessed were the metal binding properties, peptide conformations, and DNA-binding abilities of the mutants. All three mutant finger 3 peptides exhibited conformations and thermal stabilities similar to the wild-type peptide. In DNA-binding assays, the Lys mutant bound to target DNA, though its affinity was lower than that of the wild-type peptide. On the other hand, the Leu and Hgn mutants had no ability to bind DNA, despite the similarity in their secondary structures to the wild-type. Our results demonstrate that, as with the Leu residue, the aliphatic carbon side chain of this Arg residue plays a key role in the formation of a stable zinc finger domain, and its terminal guanidinium group appears to be essential for DNA binding mediated through both electrostatic interaction and hydrogen bonding with DNA phosphate backbone.

Zinc finger-zinc finger interaction between the transcription factors, GATA-1 and Sp1.

In contrast to the extensive understanding of the zinc finger-DNA interactions, less is known about zinc finger-zinc finger interactions. GATA-1 and Sp1 are transcription factors with zinc finger domains for DNA binding. The interaction between the GATA-1 and Sp1 zinc finger domains is important for synergistic transcriptional effects in erythroid genes. Despite the biological importance of the GATA-1 and Sp1 interaction, the molecular mechanism of the interaction remains unclear. We constructed a series of deletion mutants of the zinc finger domains of GATA-1 and Sp1 to identify the regions within the GATA-1 and Sp1 zinc finger domains that interact. The zinc finger-zinc finger interaction modes were also estimated from calorimetric measurements. This revealed that the interaction between the Sp1 and GATA-1 zinc finger domains was primarily electrostatic, and that the linker region of the Sp1 zinc fingers is important for the association with the GATA-1 zinc finger domains. We propose a new molecular mechanism for zinc finger-zinc finger interactions that should contribute to our understanding of the bio-functional role of the interaction between GATA-1 and Sp1.

Non-FokI-based zinc finger nucleases.

The design of functional proteins is one of the most challenging areas of protein research. We have constructed zinc finger peptides with metal-dependent hydrolytic abilities by mutating the zinc ligands in classical zinc fingers, without the need to add a FokI or other DNA cleavage domain. The designed peptides acquired DNA cleavage ability successfully, retaining the proper zinc finger folding and DNA targeting ability. We have also succeeded in site-specific DNA cleavage in the presence of cerium ions by introducing a lanthanide ion-binding loop as a linker of zinc finger motifs.

Cytochrome c polymerization by successive domain swapping at the C-terminal helix.

Cytochrome c (cyt c) is a stable protein that functions in a monomeric state as an electron donor for cytochrome c oxidase. It is also released to the cytosol when permeabilization of the mitochondrial outer membrane occurs at the early stage of apoptosis. For nearly half a century, it has been known that cyt c forms polymers, but the polymerization mechanism remains unknown. We found that cyt c forms polymers by successive domain swapping, where the C-terminal helix is displaced from its original position in the monomer and Met-heme coordination is perturbed significantly. In the crystal structures of dimeric and trimeric cyt c, the C-terminal helices are replaced by the corresponding domain of other cyt c molecules and Met80 is dissociated from the heme. The solution structures of dimeric, trimeric, and tetrameric cyt c were linear based on small-angle X-ray scattering measurements, where the trimeric linear structure shifted toward the cyclic structure by addition of PEG and (NH(4))(2)HPO(4). The absorption and CD spectra of high-order oligomers (approximately 40 mer) were similar to those of dimeric and trimeric cyt c but different from those of monomeric cyt c. For dimeric, trimeric, and tetrameric cyt c, the DeltaH of the oligomer dissociation to monomers was estimated to be about -20 kcal/mol per protomer unit, where Met-heme coordination appears to contribute largely to DeltaH. The present results suggest that cyt c polymerization occurs by successive domain swapping, which may be a common mechanism of protein polymerization.

[Artificial transcription factors based on multi-zinc finger motifs].

Artificial transcription factors targeting any desired genes are very attractive from the standpoint of regulating biological functions for life science studies and clinical applications. In order to generate such transcription factors, specific DNA binding domains are required to address a single site for each gene promoter. C(2)H(2) type zinc finger motif is one of the best frameworks to create new artificial DNA binding proteins for the following features: the zinc finger motif can recognize three bases DNA, be tandemly repeated by covalent linkage, and work as a monomer. Taking advantage of these features, manifold zinc finger proteins targeting various DNA sequences have been created so far. For application to a target in sequences as complex as the human genome, the significantly strict specificity in DNA binding must be required. Conjugating multiple fingers (multi-zinc fingers) enables to recognize longer sequences which are sufficient for addressing a single site in the human genome, whereas it has become known that as the number of finger motifs increases, the equilibrium time with the target sequence is significantly longer by in vitro experiments. Our recent study showed that the multi-zinc finger type artificial transcription factor could activate the reporter gene promptly. There is much interest in creating gene regulators, and the artificial transcription factors based on multi-zinc finger motifs could be a superior scaffold.

Binding of multivalent anionic porphyrins to V3 loop fragments of an HIV-1 envelope and their antiviral activity.

Interactions of multivalent anionic porphyrins and their iron(III) complexes with cationic peptides, V3(Ba-L) and V3(IIIB), which correspond to those of the V3 loop regions of the gp120 envelope proteins of the HIV-1(Ba-L) and HIV-1(IIIB) strains, respectively, are studied by UV/Vis, circular dichroism, (1)H NMR, and EPR spectroscopy, a microcalorimetric titration method, and anti-HIV assays. Tetrakis(3,5-dicarboxylatophenyl)porphyrin (P1), tetrakis[4-(3,5-dicarboxylatophenylmethoxy)phenyl]porphyrin (P2), and their ferric complexes (Fe(III)P1 and Fe(III)P2) were used as the multivalent anionic porphyrins. P1 and Fe(III)P1 formed stable complexes with both V3 peptides (binding constant K>10(6) M(-1)) through combined electrostatic and van der Waals interactions. Coordination of the His residues in V3(Ba-L) to the iron center of Fe(III)P1 also played an important role in the complex stabilization. As P2 and Fe(III)P2 form self-aggregates in aqueous solution even at low concentrations, detailed analysis of their interactions with the V3 peptides could not be performed. To ascertain whether the results obtained in the model system are applicable to a real biological system, anti-HIV-1(BA-L) and HIV-1(IIIB) activity of the porphyrins is examined by multiple nuclear activation of a galactosidase indicator (MAGI) and 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assays. There is little correlation between chemical analysis and actual anti-HIV activity, and the size rather than the number of the anionic groups of the porphyrin is important for anti-HIV activity. All the porphyrins show high selectivity, low cytotoxicity, and high viral activity. Fe(III)P1 and Fe(III)P2 are used for the pharmacokinetic study. Half-lives of these iron porphyrins in serum of male Wistar rats are around 4 to 6 h owing to strong interaction of these porphyrins with serum albumin.

Novel zinc finger nuclease created by combining the Cys(2)His(2)- and His(4)-type zinc finger domains.

To improve the DNA hydrolytic activity of the zinc finger nuclease, we have created a new artificial zinc finger nuclease (ZWH4) by connecting two distinct zinc finger domains possessing different types of Zn(II) binding sites (Cys(2)His(2)- and His(4)-types). The overall fold of ZWH4 is similar to that of the wild-type Sp1 zinc finger (Sp1(zf123)) as revealed by circular dichroism spectroscopy. The gel mobility shift assay demonstrated that ZWH4 binds to the GC box DNA, although the DNA-binding affinity is lower than that of Sp1(zf123). Evidently, ZWH4 hydrolyzes the covalently closed circular plasmid DNA (form I) containing the GC box (pBSGC) to the linear duplex DNA (form III) in the presence of a higher concentration (50 times) of the protein than DNA for a 24-h reaction. Of special interest is the fact that the novel mixed zinc finger protein containing the Cys(2)His(2)- and His(4)-type domains was first created. The present results provide the useful information for the redesign strategy of an artificial nuclease based on the zinc finger motif.

Effects of bulkiness and hydrophobicity of an aliphatic amino acid in the recognition helix of the GAGA zinc finger on the stability of the hydrophobic core and DNA binding affinity.

The GAGA factor of Drosophila melanogaster uses a single Cys 2His 2-type zinc finger for specific DNA binding. The conformation and DNA binding mode of the GAGA zinc finger are similar to those of other structurally characterized zinc fingers. In almost all Cys 2His 2-type zinc fingers, the fourth position of the DNA-recognizing helix is occupied by the Leu residue involved in the formation of the minimal hydrophobic core. However, no systematic study on the precise role of the Leu residue in the hydrophobic core formation and DNA binding function has been reported. In this study, the Leu residue is substituted with other aliphatic amino acids having different side chain lengths and hydrophobicities, namely, Ile, Val, Aib, and Ala. The metal binding properties were studied by UV-vis spectroscopy. The peptide conformations were examined by CD and NMR spectroscopies. Furthermore, the DNA binding ability was examined with a gel mobility shift assay. Though the Ile, Val, and Aib mutants exhibited conformations similar to those of the wild type, the DNA binding affinity decreased as the side chain length of the amino acid decreased. Interestingly, the Val mutant can bind to the cognate DNA, while Aib cannot, in spite of the similarity in their secondary structures based on the CD measurements. Variable-temperature NMR experiments clearly indicated differences in the stability of the hydrophobic core between the Val and Aib mutants. This study demonstrates that the bulkiness of the conserved aliphatic residue is important in the formation of the well-packed minimal hydrophobic core and proper ternary structure and that the hydrophobic core stabilization is apparently related to the DNA binding function of the GAGA zinc finger.

Rapid transcriptional activity in vivo and slow DNA binding in vitro by an artificial multi-zinc finger protein.

Artificial transcription factors targeting any desired genes are very attractive but require specific DNA binding domains in order to address a single site for each gene promoter. By connecting various zinc fingers recognizing the corresponding 3-4 bp DNA, DNA binding domains for the desired and long sequences can be created. Though such a long sequence recognition is a marvelous property, we have found that as the number of finger motifs increases, the equilibrium time with the target sequence is significantly longer as detected by in vitro EMSA experiments. In this study, we created 3- and 9-finger-type artificial transcription factors and compared the kinetics of the transcriptional activation in vivo as to whether or not a significant delay in the activation is observed for the 9-finger type. By using a ligand-inducing system, we demonstrated for the first time that finger multimerization does not affect the kinetics of the transcriptional activity; the 9-finger type artificial transcription factor activated the reporter gene as quickly as the 3-figner type. Our results suggest that the drawback of finger multimerization, i.e., the equilibrium time is prolonged depending on the number of finger motifs, can be surmounted in terms of its use for transcription factors in vivo. There is much interest in creating therapeutic molecules, and these findings suggest the significant potential of multi-zinc finger proteins as a tool for an artificial gene regulator.

New redesigned zinc-finger proteins: design strategy and its application.

The design of DNA-binding proteins for the specific control of the gene expression is one of the big challenges for several research laboratories in the post-genomic era. An artificial transcription factor with the desired DNA binding specificity could work as a powerful tool and drug to regulate the target gene. The zinc-finger proteins, which typically contain many fingers linked in a tandem fashion, are some of the most intensively studied DNA-binding proteins. In particular, the Cys(2)His(2)-type zinc finger is one of the most common DNA-binding motifs in eukaryotes. A simple mode of DNA recognition by the Cys(2)His(2)-type zinc-finger domain provides an ideal framework for designing proteins with new functions. Our laboratory has utilized several design strategies to create new zinc-finger peptides/proteins by redesigning the Cys(2)His(2)-type zinc-finger motif. This review focuses on the aspects of design strategies, mainly from our recent results, for the creation of artificial zinc-finger proteins, and discusses the possible application of zinc-finger technology for gene regulation and gene therapy.