Sequence selective recognition of DNA by hairpin conjugates of a racemic seco-cyclopropaneindoline-2-benzofurancarboxamide and polyamides

Conjugates of racemic seco-cyclopropaneindoline-2-benzofurancarboxamide (CI-Bf) and four diamides (ImIm 1, ImPy 2, PyIm 3, and PyPy 4, where Py is pyrrole, and Im is imidazole), linked by a gamma-aminobutyrate group were synthesized. In addition to alkylating at adenine-N3 positions within an A(5) sequence, the imidazole-containing compounds 1 and 2 were found to also alkylate purine-N3 positions within a sequence 3'-GGGGGGA(888)CTGCTC(894)-5'. A model for the binding of hairpin conjugates 1 and 2 with the 3'-GACT-5' sequence is proposed.     

DNA sequence recognition by Hoechst 33258 conjugates of hairpin pyrrole/imidazole polyamides

A series of hairpin pyrrole/imidazole polyamides linked to a Hoechst 33258 (Ht) analogue (5-7) were synthesized on solid-phase by adopting an Fmoc technique using a series of PyBOP/HOBt mediated coupling reactions. The dsDNA binding properties of Ht-polyamides 5-7 were determined by thermal denaturation experiments. Hairpin Ht-polyamides 5-7 bound to dsDNA sequences 16 and 18 show DeltaTm values that are 14-18 degrees higher than linear Ht-polyamides bound to the same sequences. All three Ht-polyamides were found to be selective for their 9-bp match dsDNA sequences, supporting a relative stronger interaction of an Im/Py anti-parallel dimer with an appropriately positioned G/Cbp rather than sequences containing only A/Tbps. In addition, Ht-polyamides 5 and 7 showed a 20-fold preference for a properly placed G/Cbp over a C/Gbp, while 6 showed a 10-fold preference.     

Sequence-specific recognition of DNA in the nucleosome by pyrrole-imidazole polyamides

The ability of DNA-binding proteins to recognize their cognate sites in chromatin is restricted by the structure and dynamics of nucleosomal DNA, and by the translational and rotational positioning of the histone octamer. Here, we use six different pyrrole-imidazole polyamides as sequence-specific molecular probes for DNA accessibility in nucleosomes. We show that sites on nucleosomal DNA facing away from the histone octamer, or even partially facing the histone octamer, are fully accessible and that nucleosomes remain fully folded upon ligand binding. Polyamides only failed to bind where sites are completely blocked by interactions with the histone octamer. Removal of the amino-terminal tails of either histone H3 or histone H4 allowed these polyamides to bind. These results demonstrate that much of the DNA in the nucleosome is freely accessible for molecular recognition in the minor groove, and also support a role for the amino-terminal tails of H3 and H4 in modulating accessibility of nucleosomal DNA.     

Sequence specific and high affinity recognition of 5'-ACGCGT-3' by rationally designed pyrrole-imidazole H-pin polyamides: thermodynamic and structural studies

Imidazole (Im) and Pyrrole (Py)-containing polyamides that can form stacked dimers can be programmed to target specific sequences in the minor groove of DNA and control gene expression. Even though various designs of polyamides have been thoroughly investigated for DNA sequence recognition, the use of H-pin polyamides (covalently cross-linked polyamides) has not received as much attention. Therefore, experiments were designed to systematically investigate the DNA recognition properties of two symmetrical H-pin polyamides composed of PyImPyIm (5) or f-ImPyIm (3e, f=formamido) tethered with an ethylene glycol linker. These compounds were created to recognize the cognate 5'-ACGCGT-3' through an overlapped and staggered binding motif, respectively. Results from DNaseI footprinting, thermal denaturation, circular dichroism, surface plasmon resonance and isothermal titration microcalorimetry studies demonstrated that both H-pin polyamides bound with higher affinity than their respective monomers. The binding affinity of formamido-containing H-pin 3e was more than a hundred times greater than that for the tetraamide H-pin 5, demonstrating the importance of having a formamido group and the staggered motif in enhancing affinity. However, compared to H-pin 3e, tetraamide H-pin 5 demonstrated superior binding preference for the cognate sequence over its non-cognates, ACCGGT and AAATTT. Data from SPR experiments yielded binding constants of 1.6x10(8)M(-1) and 2.0x10(10)M(-1) for PyImPyIm H-pin 5 and f-ImPyIm H-pin 3e, respectively. Both H-pins bound with significantly higher affinity (ca. 100-fold) than their corresponding unlinked PyImPyIm 4 and f-ImPyIm 2 counterparts. ITC analyses revealed modest enthalpies of reactions at 298 K (DeltaH of -3.3 and -1.0 kcal mol(-1) for 5 and 3e, respectively), indicating these were entropic-driven interactions. The heat capacities (DeltaC(p)) were determined to be -116 and -499 cal mol(-1)K(-1), respectively. These results are in general agreement with DeltaC(p) values determined from changes in the solvent accessible surface areas using complexes of the H-pins bound to (5'-CCACGCGTGG)(2). According to the models, the H-pins fit snugly in the minor groove and the linker comfortably holds both polyamide portions in place, with the oxygen atoms pointing into the solvent. In summary, the H-pin polyamide provides an important molecular design motif for the discovery of future generations of programmable small molecules capable of binding to target DNA sequences with high affinity and selectivity.     

Detection of CAG repeat DNA sequences by pyrene-functionalized pyrrole-imidazole polyamides

Five N-methylpyrrole-N-methylimidazole (Py-Im) polyamides possessing a fluorescent pyrene were synthesized by Fmoc solid-phase synthesis using Py/Im monomers and pyrenylbutyl-pyrrole monomer compound 9. The steady state fluorescence of conjugates 1-5 was examined in the presence and absence of (CAG)(12)-containing oligodeoxynucleotides (ODNs) 1 and 2. Of the conjugates, conjugate 1 showed no background emission around 470 nm in the absence of ODNs, and a clear increase of emission at 475 nm was observed upon addition of ODNs 1 and 2. The emission of conjugate 1 at 475 nm increased linearly with the concentration of ODN and the number of CAG repeats. The results indicate that conjugate 1 efficiently forms a pyrene excimer upon binding in the minor groove of DNA.     

Requirement of beta-alanine components in sequence-specific DNA alkylation by pyrrole-imidazole conjugates with seven-base pair recognition

To investigate the effect of incorporation of beta-alanine in alkylating N-methylpyrrole (Py)-N-methylimidazole (Im) polyamide, seco-CBI conjugates 2-8 were synthesized by an Fmoc solid-phase method and subsequent coupling with an alkylating moiety. DNA-alkylating activities of conjugates 2-8 were evaluated by high-resolution denaturing gel electrophoresis with 202-base pair (bp) DNA fragments. Alkylation by conjugates 2 and 3, which have antiparallel pairings of beta-alanine (beta) opposite beta (beta/beta) and Py/beta, occurred mainly at the adenine (A) of the matching sequences, 5'-AGCTCCA-3' (site 1) and 5'-AGCACCA-3' (site 3). However, conjugate 4, with beta/Py, did not show any DNA-alkylating activities. Similarly, conjugate 5, which possessed a Py/Py pair, weakly alkylated the matching sites at micromolar concentrations. Conjugates 6 and 7, which possessed beta/beta and Py/beta pairs, respectively, alkylated at the A of the matching sequences, 5'-ACTACCA-3' (site 2) and 5'-ACAACCA-3' (site 4). In contrast, conjugated 8, with a Py/Py pair, showed lower activity and less alkylated DNA at sites 2 and 4 with mismatched alkylation at site 1 at a higher concentration than that of 6 and 7. These results demonstrate that incorporation of beta-alanine is required for the sequence-specific alkylation by seco-CBI Py-Im conjugates with a seven-base pair sequence.     

Expanding the recognition of the minor groove of DNA by incorporation of beta-alanine in hairpin polyamides

In order to expand the recognition code by hairpin polyamides to include DNA sequences of the type 5'-CWWC-3' two polyamides, PyPyPyPy-(R)(H2N)gamma-ImPyPyIm-beta-Dp (1) and PyPyPyPy-(R)(H2N)gamma-ImPy-beta-Im-beta-Dp (2) were synthesized which have in common an Py/Im pair in the terminal position for targeting C x G but differ with respect to internal placement of a beta-alanine residue. The equilibrium association constants (Ka) were determined at four DNA sites which differ at a single common position, 5'-TNTACA-3' (N = T, A, G, C). Quantitative DNase I footprint titration experiments reveal that the eight-ring hairpin PyPyPyPy-(R)(H2N)gamma-ImPyPyIm-beta-Dp (1) binds the four binding sites with similar affinities, Ka = 1.3-1.9 x 10(10) M(-1) indicating that there is no preference for the position N. In contrast, a redesigned polyamide PyPyPyPy-(R)(H2N)gamma-ImPy-beta-Im-beta-Dp (2) that places an internal flexible aliphatic beta-alanine to the 5'-side of a key imidazole group bound the match site 5'-TCTACA-3' with high affinity and good sequence discrimination (Ka(match) = 4.9 x 10(10) M(-1) and the single base pair mismatch sites with 5- to 25-fold lower affinity). These results expand the repertoire of sequences targetable by hairpins and emphasize the importance of beta-alanine as a key element for minor groove recognition.     

Recognition of the DNA minor groove by pyrrole-imidazole polyamides: comparison of desmethyl- and N-methylpyrrole

Polyamides consisting of N-methylpyrrole (Py), N-methylimidazole (Im), and N-methyl-3-hydroxypyrrole (Hp) are synthetic ligands that recognize predetermined DNA sequences with affinities and specificities comparable to many DNA-binding proteins. As derivatives of the natural products distamycin and netropsin, Py/Im/Hp polyamides have retained the N-methyl substituent, although structural studies of polyamide:DNA complexes have not revealed an obvious function for the N-methyl. In order to assess the role of the N-methyl moiety in polyamide:DNA recognition, a new monomer, desmethylpyrrole (Ds), where the N-methyl moiety has been replaced with hydrogen, was incorporated into an eight-ring hairpin polyamide by solid-phase synthesis. MPE footprinting, affinity cleavage, and quantitative DNase I footprinting revealed that replacement of each Py residue with Ds resulted in identical binding site size and orientation and similar binding affinity for the six-base-pair (bp) target DNA sequence. Remarkably, the Ds-containing polyamide exhibited an 8-fold loss in specificity for the match site versus a mismatched DNA site, relative to the all-Py parent. Polyamides with Ds exhibit increased water solubility, which may alter the cell membrane permeability properties of the polyamide. The addition of Ds to the repertoire of available monomers may prove useful as polyamides are applied to gene regulation in vivo. However, the benefits of Ds incorporation must be balanced with a potential loss in specificity.     

Hydrophobic structure of hairpin ten-ring pyrrole-imidazole polyamides enhances tumor tissue accumulation/retention in vivo

To examine the hydrophobic structure of PI polyamides on tumor accumulation in vivo, PI polyamide-fluorescein conjugates 1–5 with the distinct number of N-methylimidazole (Im) units were synthesized. There existed an inverse relationship between the Im unit number of the compounds and their hydrophobicity. Compound 1 with one Im unit and 3 with three Im units accumulated and retained preferentially in tumor tissues compared to 5 with five Im units. These results suggest the importance of a PI polyamide’s primary structure, which partly contributes to its hydrophobic property, on its accumulation and/or retention in tumor tissues in vivo.     

Specific alkylation of human telomere repeats by hairpin pyrrole-imidazole polyamide

A novel hairpin polyamide-cyclopropapyrroloindole (CPI) conjugate PyImImIm-gamma-PyPyPyLDu86 (conjugate 11), which targets human telomere repeats d(TTAGGG)(n)/d(CCCTAA)(n), was synthesized. High resolution denaturing polyacrylamide gel electrophoresis using 44 bp DNA fragments and HPLC product analysis of a synthetic nonanucleotide demonstrated that conjugate 11 alkylates the target adenine in the telomere repeats, 5'-CCCTAA-3'. Examination of the antitumor activity of conjugate 11 using a panel of 39 cancer cell lines demonstrated that the average concentration of conjugate 11 required for 50% growth inhibition was 5.75 microM, which is superior to pepleomycin and bleomycin and comparable to cisplatin.     

Conjugation of a hairpin pyrrole-imidazole polyamide to a quinone methide for control of DNA cross-linking

A series of quinone methide precursors designed for DNA cross-linking were prepared and conjugated to a pyrrole-imidazole polyamide for selective association to the minor groove. Although reaction was only observed for DNA containing the predicted recognition sequence, yields of strand alkylation were low. Interstrand cross-linking was more efficient than alkylation but still quite modest and equivalent to that generated by a comparable conjugate containing the N-mustard chlorambucil. Varying the length of the linker connecting the polyamide and quinone methide derivative did not greatly affect the yield of DNA cross-linking. Instead, intramolecular trapping of the quinone methide intermediate by nucleophiles of the attached polyamide appears to be the major determinant that limits its reaction with DNA. Self-adducts of the quinone methide conjugate form readily and irreversibly as detected by a combination of chromatography and mass spectroscopy. This result is unlike comparable self-adducts observed for oligonucleotide conjugates that form more slowly and remain reversible. Equivalent intramolecular alkylation of a polyamide by its attached chlorambucil mustard was not observed under similar condition. The presence of DNA, however, did facilitate hydrolysis of this mustard conjugate.     

DNA sequence recognition in the minor groove by hairpin microgonotropens

Two novel microgonotropens (MGTs) comprised of hairpin N-propylaminepyrrole polyamides linked to a Hoechst 33258 (Ht) analogue (3 and 4) were synthesized on solid phase by adopting an Fmoc technique using a series of HOBt mediated coupling reactions. The dsDNA-binding properties of MGTs 3 and 4 were determined by thermal denaturation experiments. Both MGTs were found to be selective for their nine-bp match dsDNA sequence 9 and were less tolerant of G/C bp substitutions in the binding region than linear progenitor MGT 1. MGT 3 was intolerant of a G/C substitution located in the middle of the binding region and did not bind to sequences 13 and 14. MGT 4 also did not bind to sequence 13, and its linker-bound Ht moiety was found to be more sensitive to a G/C substitution in the Ht-binding target, as demonstrated by the lack of binding to sequence 16.     

Detection of triplet repeat sequences in the double-stranded DNA using pyrene-functionalized pyrrole-imidazole polyamides with rigid linkers

Methods for sequence-specific detection in double-stranded DNA (dsDNA) are becoming increasingly useful and important as diagnostic and imaging tools. Recently, we designed and synthesized pyrrole (Py)-imidazole (Im) polyamides possessing two pyrene moieties, 1, which showed an increased excimer emission in the presence of (CAG)(12)-containing oligodeoxynucleotides (ODN) 1 and 2. In this study, we synthesized bis-pyrenyl Py-Im polyamides with rigid linkers 2, 3, and 4 to improve their fluorescence properties. Among the conjugates, 2 showed a marked increase in excimer emission, which was dependent on the concentration of the target ODN and the number of CAG repeats in the dsDNA. Unlike conjugate 1, which has flexible linkers, the excimer emission intensity of 2 was retained at over 85%, even after 4h. Py-Im polyamides have the potential to be important diagnostic molecules for detecting genetic differences between individuals.     

Supercoil-dependent recognition of specific DNA sites by chromosomal protein HMG 2

The ability of the chromosomal high mobility group protein HMG 2 to recognize supercoil-dependent structures within the chicken adult beta-globin gene was investigated by examining its ability to protect such sites from digestion by S1 nuclease. Low molar ratios of HMG 2 were found to be sufficient for complete inhibition of S1 cleavage of a supercoiled plasmid containing the globin gene. Furthermore, HMG 2 protected an S1 cleavage site within the 5'-flanking region of the globin gene, in preference to a palindromic S1 site within the plasmid vector.     

Sequence diversity among related genes for recognition of specific targets in DNA molecules

Escherichia coli strains K12 and B, and a new strain designated D, each encode a characteristic restriction and modification enzyme. These enzymes (EcoK, EcoB and presumably EcoD) comprise three subunits of which one, that encoded by the so-called specificity gene (hsdS), is responsible for recognition of the DNA sequence specific to that system. The other two subunits, encoded by hsdR and hsdM, are interchangeable between systems, and the available molecular evidence suggests that the hsdR and hsdM genes are highly conserved. The DNA sequence of a segment of the hsd region that includes the hsdS gene has been determined for each of the three strains. The hsdS gene varies in length from 1335 to 1425 base-pairs and the only regions showing obvious homology, one of about 100 base-pairs and a second of about 250 base-pairs, are highly conserved. The remainder of each hsd S gene shares little, or no, homology with either of the other related specificity genes. Thus, the specificity subunits, though components of a family of closely related enzymes with very similar functions, have remarkably dissimilar primary structure.