Search

Photocontrolled transcription factors could be powerful tools for probing the roles of transcriptional processes in a variety of settings. Previously, we designed a photocontrolled DNA-binding protein based on a fusion between the bZIP region of GCN4 and photoactive yellow protein from Halorhodospira halophila [Morgan, S. A., et al. (2010) J. Mol. Biol. 399, 94-112]. Here we report a structure-based attempt to improve the degree of photoswitching observed with this chimeric protein. Using computational design tools PoPMuSiC 2.0, Rosetta, Eris, and bCIPA, we identified a series of single- and multiple-point mutations that were expected to stabilize the folded dark state of the protein and thereby enhance the degree of photoswitching. While a number of these mutations, particularly those that introduced a hydrophobic residue at position 143, did significantly enhance dark-state protein stability as judged by urea denaturation studies, dark-state stability did not correlate directly with the degree of photoswitching. Instead, the influence of mutations on the degree of photoswitching was found to be related to their effects on the degree to which DNA binding slowed the pB to pG transition in the PYP photocycle. One mutant, K143F, caused an ∼10-fold slowing of the photocycle and also showed the largest difference in the apparent K(d) for DNA binding, 3.5-fold lower, upon irradiation. This change in the apparent K(d) causes a 12-fold enhancement in the fraction bound DNA upon irradiation due to the cooperativity of DNA binding by this family of proteins. The results highlight the strengths and weaknesses of current approaches to a practical problem in protein design and suggest strategies for further improvement of designed photocontrolled transcription factors.

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.

Although many zinc finger motifs have been developed to recognize specific DNA triplets, a rational way to selectively skip a particular non-recognized gap in the DNA sequence has never been established. We have now created a 6-zinc finger peptide with an alpha-helix linker, Sp1ZF6(EAAAR)4, which selectively binds to the discontinuous recognition sites in the same phase (10 bp gap) against the opposite phase (5 bp gap) of the DNA helix. The linker peptide (EAAAR)4 forms an alpha-helix structure stabilized by salt bridges, and the helical length is estimated to be about 30 A, corresponding to that of the 10 bp DNA. The gel shift assays demonstrate that Sp1ZF6(EAAAR)4 preferably binds to the 10 bp-gapped target rather than the 5 bp-gapped target. The CD spectra show that the alpha-helical content of the (EAAAR)4 linker is higher in the complex with the 10 bp-gapped target than in the complex with the 5 bp-gapped target. The present results indicate that the alpha-helical linker is suitable for binding to the recognition sites in the same phase and that the linker induces the loss of binding affinity to the recognition sites with the opposite phase. The engineering of a helix-structured linker in the 6-zinc finger peptides should be one of the most promising approaches for selectively targeting discontinuous recognition sites depending on their phase situations.

The recent design strategy of zinc finger peptides has mainly focused on the alpha-helix region, which plays a direct role in DNA recognition. On the other hand, the study of non-DNA-contacting regions is extremely scarce. By swapping the beta-hairpin regions between the Sp1 and GLI zinc fingers, in this study, we investigated how the beta-hairpin region of the C(2)H(2)-type zinc finger peptides contributes to the DNA binding properties. Surprisingly, the Sp1 mutant with the GLI-type beta-hairpin had a higher DNA binding affinity than that of the wild-type Sp1. The result of the DNase I footprinting analyses also showed the change in the DNA binding pattern. In contrast, the GLI zinc finger completely lost DNA binding ability as a result of exchanging the beta-hairpin region. These results strongly indicate that the beta-hairpin region appears to function as a scaffold and has an important effect on the DNA binding properties of the C(2)H(2)-type zinc finger peptides.

Human immunodeficiency virus type 1 integrase catalyzes the integration of proviral DNA into the infected cell genome, so it is an important potential target for antiviral drug design. In an attempt to search for peptides that specifically interact with integrase (IN) and inhibit its function, we used an in vitro selection procedure, the phage display technique. A phage display library of random heptapeptides was used to screen for potential peptide ligands of HIV-1 IN. Several phage clones were identified that specifically bound IN. Two of the selected peptides (FHNHGKQ and HLEHLLF) exhibited a high affinity for IN and were chemically synthesized. High affinity was confirmed by a displacement assay which showed that these two synthetic peptides were able to compete with the phages expressing the corresponding peptide. These agents were assayed on the in vitro IN activities. While none of them inhibited the 3'-processing reaction, the FHNHGKQ peptide was found to be an inhibitor of the strand transfer reaction. Despite its high affinity for IN, the HLEHLLF peptide selected and assayed under the same conditions was unable to inhibit this reaction. We showed that the FHNHGKQ peptide inhibits specifically the strand transfer activity by competing with the target DNA for binding to IN. These IN-binding agents could be used as a base for developing new anti-integrase compounds as well as for structural studies of the still unknown three-dimensional structure of the entire integrase molecule.

Transforming growth factor-beta (TGF-beta) signaling regulates a wide range of cellular processes. Aberrant TGF-beta signaling has been implicated in various disease states in humans. A key element in this signaling pathway is phosphorylation of R-Smads such as Smad2 at the last two serine residues of the C-terminal sequence CSSXS (residues 463-467 in Smad2) by the TbetaRI receptor kinase. Phosphorylation results in the release of the R-Smad from the membrane-anchored protein SARA, binding to the co-mediator protein Smad4, translocation into the nucleus, and regulation of target gene expression. Expressed protein ligation was used to probe the contribution of the individual phosphate groups to Smad2 oligomerization and phosphorylation by TbetaRI. Phosphorylation at both positions was required to generate a stable homotrimer; however, the driving force for Smad2 self-association is provided by pSer465. Additionally, SARA was found to modulate the self-association of partially phosphorylated Smad2, which suggests an added role for this protein in preventing premature release of a monophosphorylated substrate from the receptor complex. In related studies, prephosphorylation of Smad2 at Ser465 was found to significantly increase the rate of phosphorylation at Ser467, suggesting that there may be specific recognition determinants within the kinase for the monophosphorylated intermediate. This information was exploited to design an improved peptide substrate for TbetaRI, which may prove valuable in the design of inhibitors of the TGF-beta pathway.

Engineered zinc finger proteins revealed that a linker sequence connecting zinc finger units has a significant effect on the DNA binding property of the protein. The recognition for a noncontiguous DNA target beyond the current recognition code of zinc finger proteins has never been determined because of the limitation of a zinc finger framework. DNA recognition of zinc finger proteins is limited only to a contiguous subset of three base pairs. We propose the recognition for a noncontiguous DNA target by inserting amino acids into the canonical linker between zinc finger units. The sequence selectivity of the new zinc finger peptides was evaluated by gel mobility shift assays. DNase I footprinting analyses clearly showed different DNA binding of various linker-extended zinc finger peptides. The application of a SPR measurement also revealed a DNA sequence selectivity of peptides. Insertion of three amino acids is enough for recognition of a noncontiguous DNA target with sequence selectivity. An extended linker will be useful for expansion of the recognition code of zinc finger proteins and for development of a new role for linker sequences in DNA binding of zinc finger proteins.

In previous studies, we have developed a technology for the rapid construction of novel DNA-binding proteins with the potential to recognize any unique site in a given genome. This technology relies on the modular assembly of modified zinc finger DNA-binding domains, each of which recognizes a three bp subsite of DNA. A complete set of 64 domains would provide comprehensive recognition of any desired DNA sequence, and new proteins could be assembled by any laboratory in a matter of hours. However, a critical parameter for this approach is the extent to which each domain functions as an independent, modular unit, without influence or dependence on its neighboring domains. We therefore examined the detailed binding behavior of several modularly assembled polydactyl zinc finger proteins. We first demonstrated that 80 modularly assembled 3-finger proteins can recognize their DNA target with very high specificity using a multitarget ELISA-based specificity assay. A more detailed analysis of DNA binding specificity for eight 3-finger proteins and two 6-finger proteins was performed using a target site selection assay. Results showed that the specificity of these proteins was as good or better than that of zinc finger proteins constructed using methods that allow for interdependency. In some cases, near perfect specificity was achieved. Complications due to target site overlap were found to be restricted to only one particular amino acid interaction (involving an aspartate in position 2 of the alpha-helix) that occurs in a minority of cases. As this is the first report of target site selection for designed, well characterized 6-finger proteins, unique insights are discussed concerning the relationship of protein length and specificity. These results have important implications for the design of proteins that can recognize extended DNA sequences, as well as provide insights into the general rules of recognition for naturally occurring zinc finger proteins.

Backbone amide hydrogen bonds play a central role in protein secondary and tertiary structure. Previous studies have shown that substitution of a backbone ester (-COO-) in place of a backbone amide (-CONH-) can selectively destabilize backbone hydrogen bonds in a protein while maintaining a similar conformation to the native backbone structure. The majority of these studies have focused on backbone substitutions that were accessible to solvent. The GCN4 coiled coil domain is an example of a stable alpha-helical dimer that possesses a well-packed hydrophobic core. Amino acids in the a and d positions of the GCN4 helix, which pack the hydrophobic core, were replaced with the corresponding alpha-hydroxy acids in the context of a chemoselectively ligated heterodimer. While the overall structure and oligomerization state of the heterodimer were maintained, the overall destabilization of the ester analogues was greater (average DeltaDeltaG of 3+ kcal mol(-1)) and more variable than previous studies. Since burial of the more hydrophobic ester should stabilize the backbone and reduce the DeltaDeltaG, the increased destabilization must come from another source. However, the observed destabilization is correlated with the protection factors for individual amide hydrogens from previous hydrogen exchange experiments. Therefore, our results suggest that backbone engineering through ester substitution is a useful approach for probing the relative strength of backbone hydrogen bonds.

The use of synthetic chemical moieties to design fully functional analogues of transcription factors will give rise to novel molecular tools for targeted gene regulation. Here we demonstrate that a synthetic molecule based on a nonpeptidic DNA-binding domain can be engineered to function as a highly potent transcription factor in vitro and in an intracellular context. The structure of this artificial transcription factor (ATF) consists of three parts: (i) triple-helix-forming oligonucleotide as a DNA-binding domain; (ii) composite linker moiety; and (iii) short synthetic peptide. The direct comparison of ATFs with natural transcription factors in in vitro assays reveals the ability of ATFs to initiate RNA transcription at the correct initiation site. In addition, the transcriptional activation potency of ATFs in vitro matches or exceeds the potency of GAL4-VP16, one of the strongest natural transcriptional activators. This remarkable biological activity is explained as a function of ATF's chemical structure. We also demonstrate for the first time that ATFs possess substantial ability to activate transcription in tissue culture cells, thus opening a prospect for practical applications in basic and applied research. The specific molecular design employed in the synthesis of ATFs may lead to the development of novel gene-targeting pharmaceuticals for treatment of fatal and chronic diseases.

We have employed a structure-based design to construct a small folding domain from the F-actin bundling protein villin that contains the amino acids necessary for the DNA binding of the basic leucine zipper protein GCN4 and have compared its DNA binding with GCN4. The monomeric motif folds into a stable domain and binds DNA in a rigid-body mechanism, while its affinity is not higher than that of the basic region peptide. The addition of the leucine zipper region to the folded domain restored its sequence-specific DNA binding comparable to that of GCN4. Unlike the monomeric folded domain, its leucine zipper derivative undergoes a conformational change upon DNA binding. CD spectral and thermodynamic studies indicate that the DNA-contacting region is folded in the presence or absence of DNA and suggest that the junction between the DNA-contacting and the leucine zipper regions transits to a helix in the presence of DNA. These results demonstrate that the structural transition outside the direct-contacting region, which adjusts the precise location of the DNA-contacting region, plays a critical role in the specific complex formation of basic leucine zipper proteins.

Basic-region leucine zipper (bZip) proteins contain a bipartite DNA-binding motif consisting of a leucine zipper dimerization domain and a basic region that directly contacts DNA. In all naturally occurring bZip proteins, the basic region is positioned N-terminal to the leucine zipper. We have designed a series of model bZip peptides in which the basic region of the yeast transcriptional activator GCN4 is placed C-terminal to its leucine zipper. DNA-binding studies demonstrate that the optimal reverse GCN4 (rGCN4) peptide is able to bind specifically and with wild-type affinity to DNA despite this unnatural arrangement of the two subdomains. These results suggest that a thermodynamic basis for the observed N-terminal positioning of the basic region relative to the dimerization domain is unlikely.

We have systematically designed and synthesized six kinds of 16-17 mer alanine-based peptides containing four to six lysine (K) and one to four asparagine (N) residues to achieve the selective binding to A.T base pairs of DNA duplexes. The position and number of K and N residues were changed in the helical structure according to common features of the DNA-binding proteins, in which K and N residues are expected to interact electrostatically with phosphate groups and to interact with A.T base pairs by hydrogen bonding, respectively. The time courses of binding of these peptides to dA(30).dT(30) and dG(30).dC(30) duplexes immobilized on a 27 MHz quartz crystal microbalance (QCM) were studied in 10 mM phosphate buffer (pH 7.5) and 40 mM NaCl at 10 degrees C. The maximum binding amounts (Deltam(max)) on a nanogram scale and binding constants (K(a)) could be obtained from the frequency decrease (mass increase) of the oligonucleotide-immobilized QCM. The conformation changes of the peptides upon binding to DNAs were monitored by circular dichroism (CD) spectroscopy. The four properly arranged N residues in the six-cationic K peptide, K6N4(d), resulted in a 5-fold higher affinity for A.T base pairs (K(a) = 5.9 x 10(5) M(-1)) than for G.C base pairs (K(a) = 1.2 x 10(5) M(-1)), and alpha-helices were clearly promoted by the binding to A.T base pairs from CD spectral changes.

Cys(2)-His(2)-type zinc finger proteins have a tandemly repeated array structure consisting of independent finger modules. They are expected to elevate the DNA binding affinity and specificity by increasing the number of finger modules. To investigate the relation between the number and the DNA binding affinity of the zinc finger, we have designed the two- to four-finger peptides by connecting the central zinc finger (finger 2) of Sp1 with the canonical linker sequence, Thr-Gly-Glu-Lys-Pro. Gel mobility shift assays reveal that the cognate three- and four-finger peptides, Sp1(zf222) and Sp1(zf2222), strongly bind to the predicted target sequences, but the two-finger peptide, Sp1(zf22), does not. Of special interest is the fact that the dissociation constant for Sp1(zf2222) binding to the target DNA is comparable to that for Sp1(zf222). The methylation interference, DNase I and hydroxyl radical footprintings, and circular permutation analyses demonstrate that Sp1(zf2222) binds to its target site with three successive zinc fingers and the binding of the fourth zinc finger is inhibited by DNA bending induced by the binding of the three-finger domain. The present results strongly indicate that the zinc finger protein binds to DNA by the three-finger domain as one binding unit. In addition, this information provides the basis for the design of a novel multifinger protein with high affinity and specificity for long DNA sequences, such as chromosomal DNAs.

The pAntp peptide, corresponding to the third helix of the homeodomain of the Antennapedia protein, enters by a receptor-independent process into eukaryotic cells. The interaction between the pAntp peptide and the phospholipid matrix of the plasma membrane seems to be the first step involved in the translocation mechanism. However, the mechanism by which the peptide translocates through the cell membrane is still not well established. We have investigated the translocation ability of pAntp through a protein-free phospholipid membrane in comparison with a more amphipathic analogue. We show by fluorescence spectroscopy, circular dichroism, NMR spectroscopy, and molecular modeling that pAntp is not sufficiently helically amphipathic to cross a phospholipid membrane of a model system. Due to its primary sequence related to its DNA binding ability in the Antennapedia homeodomain-DNA complex, the pAntp peptide does not belong to the amphipathic alpha-helical peptide family whose members are able to translocate by pore formation.

Three structural variants (PV5, PV7, and PV8) of the horseshoe crab cationic antimicrobial peptide polyphemusin I were designed with improved amphipathic profiles. Circular dichroism spectroscopy analysis indicated that in phosphate buffer polyphemusin I, PV7, and PV8 displayed the spectrum of a type II beta-turn-rich structure, but, like polyphemusin I, all three variants adopted a typical beta-sheet structure in an anionic lipid environment. Both polyphemusin I and variants were potent broad spectrum antimicrobials that were clearly bactericidal at their minimal inhibitory concentrations. The variants were moderately less active in vitro but more effective in animal models. Moreover, these variants exhibited delayed bacterial killing, whereas polyphemusin I killed Escherichia coli UB1005 within 5 min at 2.5 microg/mL. All the peptides showed similar abilities to bind to bacterial lipopolysaccharide (LPS) and permeabilize bacterial outer membranes. Consistent with this was the observation that all peptides significantly inhibited cytokine production by LPS-stimulated macrophages and penetrated polyanionic LPS monolayers to similar extents. None of the peptides had affinity for neutral lipids as evident from both tryptophan fluorescence spectroscopy and Langmuir monolayer analysis. As compared to polyphemusin I, all variants showed reduced ability to interact with anionic lipids, and the hemolytic activity of the variants was decreased by 2-4-fold. In contrast, polyphemusin I efficiently depolarized the cytoplasmic membrane of E. coli, as assessed using a membrane potential sensitive fluorescent dye 3,3-dipropylthiacarbocyanine (diSC(3)5) assay, but the variants showed a substantially delayed and decreased depolarizing ability. The coincident assessment of cell viability indicated that depolarization of the bacterial cytoplasmic membrane potential by polyphemusin I occurred prior to lethal damage to cells. Our data suggest that increase of amphipathicity of beta-sheet polyphemusin I generally resulted in variants with decreased activity for membranes. Interestingly, all variants showed an improved ability to protect mice both against infection by Pseudomonas aeruginosa and from endotoxaemia.

The basic helix-loop-helix domain of the Drosophila transcription factor Deadpan (Dpn) was prepared by total chemical protein synthesis in order to characterize its DNA binding properties. Circular dichroism spectroscopy was used to correlate structural changes in Dpn with physiologically relevant monovalent (KCl) and divalent (MgCl2) cation concentrations. In addition, we have used electrophoretic mobility shift assay (EMSA) and fluorescence anisotropy methods to determine equilibrium dissociation constants for the interaction of Dpn with two biologically relevant promoters involved in neural development and sex determination pathways. In this study, we have optimized DNA binding conditions for Dpn, and we have found a markedly higher DNA binding affinity for Dpn than reported for other bHLH domain transcription factors. Dpn binds as a homodimer (Kd = 2.6 nM) to double-stranded oligonucleotides containing the binding site CACGCG. In addition, we found that Dpn bound with the same affinity to a single or tandem binding site, indicating no cooperativity between adjacent DNA-bound Dpn dimers. DNA binding was also monitored as a function of physiologically relevant KCl and MgCl2 concentrations, and we found that this activity was significantly different in the presence and absence of the nonspecific competitor poly(dI-dC). Moreover, Dpn displayed moderate sequence selectivity, exhibiting a 100-fold higher binding affinity for specific DNA than for poly(dI-dC). This study constitutes the first detailed biophysical characterization of the DNA binding properties of a bHLH protein.

Sequence-specific DNA binding proteins generally consist of more than two DNA-contacting regions to ensure the selectivity of recognition. The multiple DNA binding modules are connected either through the covalent linker or through the noncovalent dimerization domain. We have compared the DNA binding of peptide dimers with covalent and noncovalent dimerization domains to explore the potential advantage of each linkage on the sequence-specific DNA binding. Three sets of head-to-tail peptide dimers were synthesized by using the same basic region peptide to target the same DNA sequence; one dimer was assembled with a bridged biphenyl derivative as a covalent dimerization domain, and two other dimers were assembled with the cyclodextrin guest noncovalent dimerization domains. One of the noncovalent dimers was a heterodimer that consisted of cyclodextrin and guest peptides, while the other was a homodimer that consisted of peptides bearing both cyclodextrin and the guest molecule within the same chain. Both noncovalent dimers formed the specific DNA complexes within narrower ranges of peptide concentrations and showed higher sequence selectivity than the covalent dimer did. Among the three dimers, the noncovalent homodimer that can form an intramolecular inclusion complex showed the highest sequence selectivity. Because the noncovalent homodimer with the higher stability of the circular intramolecular inclusion complex exhibited the higher sequence selectivity, it was concluded that an equilibrium involving a conformational transition of a monomeric peptide effectively reduced the stability of its nonspecific binding complex, hence increasing the efficacy of cooperative dimer formation at the specific DNA sequence.

Terminase is an enzyme from bacteriophage lambda that is required for insertion of the viral genome into an empty pro-capsid. This enzyme is composed of the viral proteins gpNu1 (20.4 kDa) and gpA (73.3 kDa) in a holoenzyme complex. Current models for terminase assembly onto DNA suggest that gpNu1 binds to three repeating elements within a region of the lambda genome known as cosB which, in turn, stimulates the assembly of a gpA dimer at the cosN subsite. This prenicking complex is the first of several stable nucleoprotein intermediates required for DNA packaging. We have noted a hydrophobic region within the primary amino acid sequence of the terminase gpNu1 subunit and hypothesized that this region constitutes a protein-protein interaction domain required for cooperative assembly at cosB and that is also responsible for the observed aggregation behavior of the isolated protein. We therefore constructed a mutant of gpNu1 in which this hydrophobic "domain" has been deleted in order to test these hypotheses. The deletion mutant protein, gpNu1DeltaK, is fully soluble and, unlike full-length protein, shows no tendency toward aggregation; However, the protein is a dimer under all experimental conditions examined as determined by gel permeation and sedimentation equilibrium analysis. The truncated protein is folded with evidence of secondary and tertiary structural elements by circular dichroism and NMR spectroscopy. While physical and biological assays demonstrate that gpNu1DeltaK does not interact with the terminase gpA subunit, the deletion mutant binds with specificity to cos-containing DNA. We have thus constructed a deletion mutant of the phage lambda terminase gpNu1 subunit which constitutes a highly soluble DNA binding domain of the protein. We further propose that the hydrophobic amino acids found between Lys100 and Pro141 define a self-association domain that is required for the assembly of stable nucleoprotein packaging complexes and that the C-terminal tail of the protein defines a distinct gpA-binding site that is responsible for terminase holoenzyme formation.

It is well-known that DNA binding of native nine zinc-finger protein TEIIIA is dominated by interaction of select few fingers. Newly designed zinc-finger peptide Sp1ZF9 containing nine Cys2-His2 type motifs has been manipulated. The DNA-binding property of Sp1ZF9 was compared with those of native three zinc-finger Sp1(530-623) and artificial six zinc-finger Sp1ZF6 peptides. Although the equilibrium time was less than 0.5 h for Sp1(530-623)-DNA complex, Sp1ZF6 and Sp1ZF9 required approximately 48 and 72 h, respectively, for full complex formation. Evidently, the footprinting analysis demonstrated that Sp1ZF9 and Sp1ZF6 bind at least 27 and 18 contiguous base pairs of DNA sequence, respectively. Sp1ZF9 showed two step bindings to DNA, namely first the recognition of GC (5'-GGG-GCG-GGGCC-3') sequence by the N-terminal Sp1 domain and next the recognition of the corresponding target sequences by the middle and C-terminal Sp1 domains. In contrast with unimolecular binding of Sp1ZF9 and Sp1ZF6, two Sp1(530-623) molecules bind to one GCIII (5'-GGG-GCG-GGG-GGG-GCG-GGG-GGG++ +-GCG-GGGCC-3') site region. Semispecific complex formed at the beginning of Sp1ZF9-DNA interaction has also been characterized by kinetic analysis using surface plasmon resonance. Interestingly, the association rate constants for GC and GCIII complexes of Sp1ZF9 are smaller than those of the corresponding Sp1(530-623) complexes. Of special interest is the fact that new nine zinc-finger peptide Sp1ZF9 can bind to DNA sequence of approximately 30 base pairs. Such multi zinc-finger peptides may be useful as genome-specific transcriptional switches in future.

The overall affinity of the bacteriophage lambda Cro repressor for its operator DNA site is limited by dimer dissociation at submicromolar concentrations. Since Cro dimer-operator complexes form at nanomolar concentrations of Cro subunits where free dimers are rare, these dimers must bind with compensating high affinities. Previous studies of the covalent dimer Cro V55C suggest little change in DNA binding affinity even though the dimeric species is quantitatively populated; this is an apparent contradiction to the expectation of high intrinsic dimer-DNA affinity. In contrast to the disulfide linkage at the center of the dimer interface in Cro V55C, polypeptide linkers that join the two subunits allow single-chain Cro repressors to bind operator DNA with picomolar affinities. A series of five single-chain Cro repressors have been expressed from fused tandem cro genes. Each contains a peptide linker of 8-16 hydrophilic residues that connects the C-terminus of one subunit to the N-terminus of the next. All bind to operator DNA with at least 100-fold higher affinity than Cro V55C. Proteins containing the longest and shortest linkers have been purified and characterized in detail. Both exhibit similar CD spectra to wild-type Cro and enhanced thermal stability. Sedimentation equilibrium experiments show that single-chain Cro repressors do not associate at concentrations up to 30 microM. The rate of dissociation of Cro-DNA complexes is almost unchanged by covalent linkage. Biophysical characterization of Cro variants such as these, where DNA binding is uncoupled from subunit assembly, is necessary for a quantitative understanding of the structural and energetic determinants of DNA recognition in this simple model system.

We have designed a cationic amphipathic peptide, KALA (WEAKLAKALAKALAKHLAKALAKALKACEA), that binds to DNA, destabilizes membranes, and mediates DNA transfection. KALA undergoes a pH-dependent random coil to amphipathic alpha-helical conformational change as the pH is increased from 5.0 to 7.5. One face displays hydrophobic leucine residues, and the opposite face displays hydrophilic lysine residues. KALA-mediated release of entrapped aqueous contents from neutral and negatively charged liposomes increases with increasing helical content. KALA binds to oligonucleotides or plasmid DNA and retards their migration in gel electrophoresis. It displaces 50% of ethidium bromide from DNA at a charge ratio (+/-) of 0.9/1. In cultured cells, KALA assists oligonucleotide nuclear delivery when complexes are prepared at a 10/1 (+/-) charge ratio. KALA/DNA (10/1)(+/-) complexes mediate transfection of a variety of cell lines. The KALA sequence provides a starting point for a family of peptides that incorporate other functions to improve DNA delivery systems.

A 55-residue protein containing the DNA binding domain of Hin recombinase, residues 139-190, with the tripeptide Gly-Gly-His (GGH) at the NH2 terminus was synthesized by stepwise solid-phase methods. GGH(Hin139-190) binds sequence specifically to DNA at four 13 base pair sites (termed hixL and secondary) and, in the presence of Ni(OAc)2 and monoperoxyphthalic acid, reacts predominantly at a single deoxyribose position on one strand of each binding site [Mack, D.P., & Dervan, P.B. (1990) J. Am. Chem. Soc. 112, 4604]. We find that, upon treatment with n-butylamine, the DNA termini at the cleavage site are 3'- and 5'-phosphate, consistent with oxidative degradation of the deoxyribose backbone. The nickel-mediated oxidation can be activated with peracid, iodosylbenzene, or hydrogen peroxide. The sequence specificity of the reaction is not dependent on oxidant, but the rates of cleavage differ, decreasing in the order peracid greater than iodosylbenzene greater than hydrogen peroxide. Optimal cleavage conditions for a 1 microM concentration of protein are 50 microM peracid, pH 8.0, and 1 equiv of Ni(OAc)2. The preferential cleavage at a single base pair position on one strand of the minor groove indicates a nondiffusible oxidizing species. A change of absolute configuration in the GGH metal binding domain from L-His to D-His [Ni(II).GG-(-D-)H(Hin139-190)] affords cleavage at similar base pair locations but opposite with regard to strand specificity.