Differences in sequence selectivity of DNA alkylation by isomeric intercalating aniline mustards

Two DNA-targeted mustard derivatives, N,N-bis(2-chloroethyl)-4-(5-[9-acridinylamino]-pentamido)aniline and 4-(9-[acridinylamino]butyl 4-(N,N-bis[2-chloroethyl]-aminobenzamide, which are isomeric compounds where the mustard is linked to the DNA-binding 9-aminoacridine moiety by either a -CONH- or a -NHCO- group, show significant differences in the sequence selectivity of their alkylation of DNA. The CONH isomer is a more efficient alxylating agent than the NHCO compound by an order of magnitude, consistent with the larger electron release of the CONH group to the aniline ring. However, the pattern of alkylation by the two compounds is also very different, with the CONH isomer preferring alkylation of guanines adjacent to 3'- or 5'-adenines and cytosines (for example those in sequences 5'-CGC, 5'-AGC, 5'-CGG and 5'-AGA) while the isomeric NHCO compound shows preference for guanines in runs of Gs. In addition, both isomers alkylate 3'-adenines in runs of adenines. Both compounds also show completely different patterns of alkylation to their untargeted mustard counterparts, since 4-MeCONH-aniline mustard alkylates all guanines and adenines in runs of adenines, while 4-Me2NCO-aniline mustard fails to alkylate DNA at all. These differences in alkylation patterns between the CONH- and its isomeric NHCO- compounds and their relationships between the alkylation patterns of the isomers and their biological activities are discussed.     

DNA stretching and extreme kinking in the nucleosome core

DNA stretching in chromatin may facilitate its compaction and influence site recognition by nuclear factors. In vivo, stretching has been estimated to occur at the equivalent of one to two base-pairs (bp) per nucleosome. We have determined the crystal structure of a nucleosome core particle containing 145 bp of DNA (NCP145). Compared to the structure with 147 bp, the NCP145 displays two incidences of stretching one to two double-helical turns from the particle dyad axis. The stretching illustrates clearly a mechanism for shifting DNA position by displacement of a single base-pair while maintaining nearly identical histone-DNA interactions. Increased DNA twist localized to a short section between adjacent histone-DNA binding sites advances the rotational setting, while a translational component involves DNA kinking at a flanking region that initiates elongation by unstacking bases. Furthermore, one stretched region of the NCP145 displays an extraordinary 55° kink into the minor groove situated 1.5 double-helical turns from the particle dyad axis, a hot spot for gene insertion by HIV-integrase, which prefers highly distorted substrate. This suggests that nucleosome position and context within chromatin could promote extreme DNA kinking that may influence genomic processes.     

Inhibition of simian virus 40 DNA replication in CV-1 cells by an oligodeoxynucleotide covalently linked to an intercalating agent.

An octathymidylate covalently linked via its 3'-end to an acridine derivative inhibited the cytopathic effect of Simian Virus SV40 on CV-1 cells in culture. Control experiments revealed that this effect was virus-specific and did not arise as a result of oligonucleotide degradation by nucleases. A photoactive probe was covalently attached to the 5'-end of the oligonucleotide-acridine conjugate. Upon UV-irradiation, photocrosslinking was shown to occur at the A. T-rich region within the viral origin of replication. A local triple helix can form at moderate salt concentrations with two octathymidylate-acridine conjugates bound to the octaadenylate sequence. Alternatively the octathymidylate-acridine conjugate can bind to the major groove of duplex DNA forming a local triple helix. Different mechanisms are discussed to explain the inhibition of viral DNA replication.     

The DNA chaperone HMGB1 facilitates ACF/CHRAC-dependent nucleosome sliding

Nucleosome remodelling complexes CHRAC and ACF contribute to chromatin dynamics by converting chemical energy into sliding of histone octamers on DNA. Their shared ATPase subunit ISWI binds DNA at the sites of entry into the nucleosome. A prevalent model assumes that DNA distortions catalysed by ISWI are converted into relocation of DNA relative to a histone octamer. HMGB1, one of the most abundant nuclear non-histone proteins, binds with preference to distorted DNA. We have now found that transient interaction of HMGB1 with nucleosomal linker DNA overlapping ISWI-binding sites enhances the ability of ACF to bind nucleosomal DNA and accelerates the sliding activity of limiting concentrations of remodelling factor. By contrast, an HMGB1 mutant with increased binding affinity was inhibitory. These observations are consistent with a role for HMGB1 as a DNA chaperone facilitating the rate-limiting DNA distortion during nucleosome remodelling.     

DNA alkylation and formation of DNA interstrand cross-links by potential antitumour 2,5-bis(1-aziridinyl)-1,4-benzoquinones

A series of 3,6-substituted 2,5-bis(1-aziridinyl)-1,4-benzoquinone derivatives was shown to alkylate calf thymus DNA and to form DNA interstrand cross-links. Alkylation and cross-link formation were enhanced after electrochemical reduction of the compounds and increased with lower pH in the pH range from 4.5 to 8.0. Reduction especially shifts the pH at which cross-linking and alkylation occurs to higher values, which are more physiologically relevant. This shift is probably caused by the increase in pK_a value of the aziridine ring after reduction of the quinone moiety. The inactivation of single-stranded bacteriophage M13mp19 DNA to form phages in an E. coli host, by the 3,6-unsubstituted parent compound 2,5-bis(1-aziridinyl)-1,4-benzoquinone (TW13) was dependent upon reduction and pH in a similar way as was alkylation. The compound in our series with the least bulky, 3,6-substitutents, TW13, caused a high amount of cross-link formation. Compounds with methyl-substituted aziridine rings showed low cross-linking ability. Our results support the concept that the protonated reduced compound is the reactive species that alkylates DNA, and that steric factors play an important role in the reactivity towards DNA. A correlation is observed between the ability to induce DNA interstrand cross-links and inactivation of M13mp19 bacteriophage DNA. Cross-link formation was also demonstrated in E. coli K12 cells, where the compounds are reduced endogenously by bacterial reductases.     

Analysis of DNA intercalating drugs by TGGE

Temperature-gradient gel electrophoresis (TGGE) was used to study DNA-drug interactions. The results indicate that at least two classes of DNA intercalating drugs are distinguishable with respect to temperature increase: reversible and irreversible. The method offers an excellent means of visualizing the melting profile of an individual DNA topoisomer in the presence of DNA binding drugs. Our findings coincide with UV/VIS absorption spectroscopy data.     

DNA sequence preferences for an intercalating porphyrin compound revealed by footprinting.

The DNA sequence preferences of the compound meso-tetra-(4-N-methyl(pyridyl) porphyrin and its nickel complex have been investigated by means of footprinting experiments on several DNA fragments, using DNAase I and micrococcal nuclease as footprinting agents. A complex pattern of both AT and GC-protected sites was found. Ligand-induced long-range conformational changes were inferred in several instances to be related to the observed large-scale blockages of enzymatic cutting.     

DNA folding in the nucleosome

Digestion of chromatin with a number of nucleases shows that the DNA is regularly folded in the nucleosome. Particularly cleavage by pancreatic DNase (DNase I) in the 140 base-pair nucleosome has been examined. This nuclease nicks the DNA every ten bases on each strand as demonstrated by labeling the 5′-ends of the 140 base-pair nucleosome. Cleavage sites on opposite strands are staggered by two bases. This proves that the DNA is arranged on the outside of the histone core in a regular way. The probability distribution of nicking might indicate a 2-fold symmetry of the 140 base-pair nucleosome. In particular it is shown that the predominant band of 80 bases is derived from several regions within the 140 base-pairs and suggested to reflect the pitch of the DNA superhelix surrounding the histone core of the nucleosome. Its possible significance with respect to chromatin structure is discussed.     

Targeting radiosensitizers to DNA by attachment of an intercalating group: nitroimidazole-linked phenanthridines

The nitroimidazole-linked phenanthridine series of compounds (NLP-1, 2, and 3) were synthesized under the assumption that it should be possible to enhance the molar efficiency of 2-nitroimidazoles as hypoxic cell radiosensitizers and cytotoxins by targeting them to their likely site of action, DNA. The targeting group chosen was the phenanthridine moiety, the major component of the classical DNA intercalating compound, ethidium bromide. The sole difference between the compounds is the length of the hydrocarbon chain linking the nitroimidazole to the phenanthridine. The phenanthridine group with a three-carbon side chain, P-1, was also synthesized to allow studies on the effect of the targeting group by itself. The ability of the compounds to bind to DNA is inversely proportional to their linker chain length with binding constant values ranging from approximately 1 x 10(5) mol-1 for NLP-2 to 6 x 10(5) mol-1 for NLP-3. The NLP compounds show selective toxicity to hypoxic cells at 37 degrees C at external drug concentrations 10-40 times lower than would be required for untargeted 2-nitroimidazoles such as misonidazole in vitro. Toxicity to both hypoxic and aerobic cells is dependent on the linker chain: the shorter the chain, the greater the toxicity. In addition, the NLP compounds radiosensitize hypoxic cells at external drug concentrations as low as 0.05 mM with almost the full oxygen effect being observed at a concentration of 0.5 mM. These concentrations are 10-100 times lower than would be required for similar radiosensitization using misonidazole. Radiosensitizing ability is independent of linker chain length. The present compounds represent prototypes for further studies of the efficacy and mechanism of action of 2-nitroimidazoles targeted to DNA by linkage to an intercalating group.     

Inhibition of cation-induced DNA condensation by intercalating dyes

Several intercalating dyes are shown to inhibit the cation-induced condensation of λ-DNA when Co³⁺(NH₃)₆ is the condensing agent. The dyes that have been studied are ethidium, propidium, proflavin, quinacrine, and actinomycin D. Earlier work has shown that intercalating dyes inhibit ψ-DNA condensation. [Lerman, L. S. (1971) Prog. Mol. Subcell. Biol. 2 , 382–391; Cheng, S. & Mohr, S. C. (1975) Biopolymers 14 , 663–674.] Dye-induced decondensation of intramolecularly condensed DNA has been studied by making use of conditions in which Co³⁺(NH₃)₆ produces intramolecular condensation without significant aggregation. Some aggregation is caused, however, during dye-induced decondensation. Dye titration curves of DNA decondensation have been measured by excess light scattering to monitor decondensation and by fluorescence to monitor intercalation. All of the dyes studied act as competing cations in displacing the condensing cation Co³⁺(NH₃)₆ from the DNA. Competition occurs both in and below the transition zone for condensation. The effectiveness of a dye as a competing cation increases with its net positive charge. Before decondensation begins, no intercalated dye can be detected, suggesting that intercalation might be incompatible with the proper helix packing needed for cation-induced DNA condensation. To test this last point, methidium–spermine was synthesized: it contains an intercalating methidium head group combined with a polyamine tail. Methidium–spermine is found to cause λ-DNA condensation, but aggregation accompanies condensation, as has been found earlier for spermine and spermidine. Fluorescence and absorption spectra indicate that the methidium group is intercalated when the DNA is condensed, indicating that intercalation need not be incompatible with DNA condensation. The presence of aggregates among the condensed DNA molecules makes this last conclusion tentative.     

Role of recombinational repair in sensitivity to an antitumour agent that inhibits bacteriophage T4 type II DNA topoisomerase

The bacteriophage T4-encoded type II DNA topoisomerase is the major target for the antitumour agent m-AMSA (4-(9-acridinylamino)methanesulphon-m-anisidide) in phage-infected bacterial cells. Inhibition of the purified enzyme by m-AMSA results in formation of a cleavage complex that contains the enzyme covalently attached to DNA on both sides of a double-strand break. In this article, we provide evidence that this cleavage complex is responsible for inhibition of phage growth and that recombinational repair can reduce sensitivity to the antitumour agent, presumably by eliminating the complex (or some derivative thereof). First, topoisomerase-deficient mutants were shown to be resistant to m-AMSA, indicating that m-AMSA inhibits growth by inducing the cleavage complex rather than by inhibiting enzyme activity. Second, mutations in several phage genes that encode recombination proteins (uvsX, uvsY, 46 and 59) increased the sensitivity of phage T4 to m-AMSA, strongly suggesting that recombination participates in the repair of topoisomerase-mediated damage. Third, m-AMSA stimulated recombination in phage-infected bacterial cells, as would be expected from the recombinational repair of DNA damage. Finally, m-AMSA induced the production of cleavage complexes involving the T4 topoisomerase within phage-infected cells.     

Selective strand scission by intercalating drugs at DNA bulges

A bulge is an extra, unpaired nucleotide on one strand of a DNA double helix. This paper describes bulge-specific strand scission by the DNA intercalating/cleaving drugs neocarzinostatin chromophore (NCS-C), bleomycin (BLM), and methidiumpropyl-EDTA (MPE). For this study we have constructed a series of 5'-32P end labeled oligonucleotide duplexes that are identical except for the location of a bulge. In each successive duplex of the series, a bulge has been shifted stepwise up (from 5' to 3') one strand of the duplex. Similarly, in each successive duplex of the series, sites of bulge-specific scission and protection were observed to shift in a stepwise manner. The results show that throughout the series of bulged duplexes NCS-C causes specific scission at a site near a bulge, BLM causes specific scission at a site near a bulge, and MPE-Fe(II) causes specific scission centered around the bulge. In some sequences, NCS-C and BLM each cause bulge-specific scission at second sites. Further, bulged DNA shows sites of protection from NCS-C and BLM scission. The results are consistent with a model of bulged DNA with (1) a high-stability intercalation site at the bulge, (2) in some sequences, a second high-stability intercalation site adjacent to the first site, and (3) two sites of relatively unstable intercalation that flank the two stable intercalation sites. On the basis of our results, we propose a new model of the BLM/DNA complex with the site of intercalation on the 3' side (not in the center) of the dinucleotide that determines BLM binding specificity.(ABSTRACT TRUNCATED AT 250 WORDS)     

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.     

Efficient DNA alkylation by a pyrrole-imidazole CBI conjugate with an indole linker: sequence-specific alkylation with nine-base-pair recognition

Conjugates 7, 8, and 10 of N-methylpyrrole (Py)-N-methylimidazole (Im) polyamides and 1,2,9,9a-tetrahydrocyclopropa[1,2-c]benz[1,2-e]indol-4-one (CBI) with a 5-amino-1H-indole-2-carbonyl linker were synthesized by Fmoc solid-phase synthesis and a subsequent liquid-phase coupling procedure. The DNA alkylating abilities of conjugates 7, 8, 6b, and 10 were examined using Texas Red-labeled PCR fragments and high-resolution denaturing gel electrophoresis. CBI conjugates 7 and 8 exhibited highly efficient sequence-specific DNA alkylation comparable with previous CBI conjugates with a vinyl linker. In particular, conjugate 10, with a 10-ringed hairpin Py-Im polyamide, alkylated at the adenine of 5'-ACAAATCCA-3'. Introduction of an indole linker greatly facilitated the synthesis of sequence-specific alkylating Py-Im polyamides.     

Dynamics of nucleosome invasion by DNA binding proteins

Nucleosomes sterically occlude their wrapped DNA from interacting with many large protein complexes. How proteins gain access to nucleosomal DNA target sites in vivo is not known. Outer stretches of nucleosomal DNA spontaneously unwrap and rewrap with high frequency, providing rapid and efficient access to regulatory DNA target sites located there; however, rates for access to the nucleosome interior have not been measured. Here we show that for a selected high-affinity nucleosome positioning sequence, the spontaneous DNA unwrapping rate decreases dramatically with distance inside the nucleosome. The rewrapping rate also decreases, but only slightly. Our results explain the previously known strong position dependence on the equilibrium accessibility of nucleosomal DNA, which is characteristic of both selected and natural sequences. Our results point to slow nucleosome conformational fluctuations as a potential source of cell-cell variability in gene activation dynamics, and they reveal the dominant kinetic path by which multiple DNA binding proteins cooperatively invade a nucleosome.     

DNA-directed alkylating ligands as potential antitumor agents: sequence specificity of alkylation by intercalating aniline mustards

The sequence preferences for alkylation of a series of novel parasubstituted aniline mustards linked to the DNA-intercalating chromophore 9-aminoacridine by an alkyl chain of variable length were studied by using procedures analogous to Maxam-Gilbert reactions. The compounds alkylate DNA at both guanine and adenine sites. For mustards linked to the acridine by a short alkyl chain through a para O- or S-link group, 5'-GT sequences are the most preferred sites at which N7-guanine alkylation occurs. For analogues with longer chain lengths, the preference of 5'-GT sequences diminishes in favor of N7-adenine alkylation at the complementary 5'-AC sequence. Magnesium ions are shown to selectively inhibit alkylation at the N7 of adenine (in the major groove) by these compounds but not the alkylation at the N3 of adenine (in the minor groove) by the antitumor antibiotic CC-1065. Effects of chromophore variation were also studied by using aniline mustards linked to quinazoline and sterically hindered tert-butyl-9-aminoacridine chromophores. The results demonstrate that in this series of DNA-directed mustards the noncovalent interactions of the carrier chromophores with DNA significantly modify the sequence selectivity of alkylation by the mustard. Relationships between the DNA alkylation patterns of these compounds and their biological activities are discussed.