Prediction of DNA-binding specificity in zinc finger proteins

Zinc finger proteins interact via their individual fingers to three base pair subsites on the target DNA. The four key residue positions -1, 2, 3 and 6 on the alpha-helix of the zinc fingers have hydrogen bond interactions with the DNA. Mutating these key residues enables generation of a plethora of combinatorial possibilities that can bind to any DNA stretch of interest. Exploiting the binding specificity and affinity of the interaction between the zinc fingers and the respective DNA can help to generate engineered zinc fingers for therapeutic purposes involving genome targeting. Exploring the structure-function relationships of the existing zinc finger-DNA complexes can aid in predicting the probable zinc fingers that could bind to any target DNA. Computational tools ease the prediction of such engineered zinc fingers by effectively utilizing information from the available experimental data. A study of literature reveals many approaches for predicting DNA-binding specificity in zinc finger proteins. However, an alternative approach that looks into the physico-chemical properties of these complexes would do away with the difficulties of designing unbiased zinc fingers with the desired affinity and specificity. We present a physico-chemical approach that exploits the relative strengths of hydrogen bonding between the target DNA and all combinatorially possible zinc fingers to select the most optimum zinc finger protein candidate.     

Binding specificity of the two major DNA-binding proteins in human serum.

The two major DNA-binding proteins of human serum (DNA-binding protein 1 and DNA-binding protein 2) were shown to bind preferentially to single-stranded polynucleotides rich in guanine residues. Equilibrium competition experiments using a nitrocellulose filter assay system containing labeled human lymphocyte DNA and various competing natural and synthetic polynucleotides indicated that both proteins recognized sequences of bases containing a keto group in either position 6 (purines) or 4 (pyrimidines) and that these keto groups must be readily accessible for effective binding to occur. Guanine was shown to be the preferred nucleotide through inhibition experiments using a series of synthetic homopolymers and a series of bacterial DNAs of differing G + C content. The relationship between protein affinity and G + C content was shown to be directly proportional. The equilibrium constants for the binding of the human lymphocyte DNA by both proteins were on the order of 10(-6) M, and the length of the nucleotide sequence necessary for effective binding was found to be 12 to 18 bases using a series of oligomers of poly(dG).     

Shuttling of SOX proteins

The control of access of SOX proteins to their nuclear target genes is a powerful strategy to activate or repress complex genetic programs. The sub-cellular targeting sequences of SOX proteins are concentrated within the DNA binding motif, the HMG (for high mobility group) domain. Each SOX protein displays two different nuclear localization signals located at the N-terminal and C-terminal part of their highly conserved DNA binding domain. The N-terminal nuclear localization signal binds calmodulin and is potentially regulated by intracellular calcium signalling, while the C-terminal nuclear localization signal, which binds importin-β, responds to other signalling pathways such as cyclic AMP/protein kinase A. Mutations inducing developmental disorders like sex reversal have been reported in both NLSs of SRY, interfering with its nuclear localization and suggesting that both functional nuclear localization signal are required for its nuclear activity. A nuclear export signal is also present in the HMG box of SOX proteins. Group E SOX proteins harbour a perfect consensus nuclear export signal sequence in contrast to all other SOX proteins, which display only imperfect ones. However, observations made during mouse embryonic development suggest that non-group E SOX proteins could also be regulated by a nuclear export mechanism. The presence of nuclear localization and nuclear export signal sequences confers nucleocytoplasmic shuttling properties to SOX proteins, and suggests that cellular events regulated by SOX proteins are highly dynamic.     

DNA-binding specificity of the S8 homeodomain.

The murine S8 homeobox gene is expressed in a mesenchyme-specific pattern in embryos, as well as in mesodermal cell lines. The S8 homeodomain is overall similar to paired type homeodomains, but at position 50, which is crucial for specific DNA recognition, it contains a Gln, as is found in Antennapedia (Antp)-type homeodomains. We determined the DNA-binding specificity of the purified S8 homeodomain by in vitro selection of random oligonucleotides. The resulting 11-bp consensus binding site, ANC/TC/TAATTAA/GC resembles, but subtly differs from, the recognition sequences of Antp-type homeodomains. Equilibrium binding constants of down to 6.0 x 10(-10) M were found for binding of the S8 homeodomain to selected oligonucleotides. Using specific antibodies and an oligonucleotide containing an S8-site, we detected by band-shift two abundant DNA binding activities in mesodermal cell lines that correspond to S8 and two more that correspond to its close relative MHox. These S8 protein forms are differentially expressed in retinoic acid-treated P19 EC cells.     

The detection of DNA-binding proteins by protein blotting.

A method, called "protein blotting," for the detection of DNA-binding proteins is described. Proteins are separated on an SDA-polyacrylamide gel. The gel is sandwiched between 2 nitrocellulose filters and the proteins allowed to diffuse out of the gel and onto the filters. The proteins are tightly bound to each filter, producing a replica of the original gel pattern. The replica is used to detect DNA-binding proteins, RNA-binding proteins or histone-binding proteins by incubation of the filter with [32P]DNA, [125I]RNA, or [125I] histone. Evidence is also presented that specific protein-DNA interactions may be detected by this technique; under appropriate conditions, the lac repressor binds only to DNA containing the lac operator. Strategies for the detection of specific protein-DNA interactions are discussed.     

The base-pairing specificity of cellulose-pdT9.

The stability of the interaction of oligoadenylates, containing single nucleotide substitutions, with cellulose-pdT9 has been studied by thermal elution. In the case of oligodeoxyriboadenylates, the replacement of an internal dA by dC, dG or dT caused destabilization. In the case of oligoriboadenylates, replacement of an internal A residue by C or U resulted in a significantly lesser destabilization. The results indicate that caution should be exercised in extrapolation of data on the specificity of deoxyribonpolynucleotide-deoxyribopolynucleotide hybridization to structure of the type deoxyribopolynucleotide-ribopolynucleotide.     

DNA-binding activity of papillomavirus proteins.

We demonstrate DNA binding by papillomavirus (PV) open reading frame (ORF) proteins that correspond to the early transforming and trans-activating (E6 and E2) and late structural regions (L2 and L1) from bovine PV type 1 and human PV types 6b and 16. All PV proteins were synthesized in Escherichia coli and had a common 13-amino-acid leader sequence from the expression vector pRA10. Antibodies have been generated in rabbits against these PV proteins. The PV ORF proteins bind double-stranded DNA, and this activity is demonstrated to be inherent to the PV proteins. DNA-binding activity by PV proteins is optimal at 50 mM NaCl and at pH 7.0. For some PV proteins (e.g., bovine PV type 1 E2), DNA binding is enhanced at a lower pH (pH 6.0) and NaCl concentration (50 to 100 mM). DNA binding is inhibited by the appropriate antibodies. The possible significance of these findings is discussed in relation to the genetic and structural evidence on the function of these ORFs.