Origins of the specificity in the intercalation of ethidium into nucleic acids. A theoretical analysis

Molecular dynamic simulations of diacridine binding to DNA: Indications that C6 diacridine can bisintercalate spanning two base pairs

Dimers of 9-aminoacridine linked via the 9-amino group with polymethylene chains, termed diacridines, are known to bisintercalate into DNA when the linker comprises 6 or more methylene units. There are no literature reports of crystal or NMR solution structures for bisintercalated diacridine-DNA complexes, and the issue of the structure of the C6 ([CH₂ ]_n linker where n = 6) diacridine complex remains unresolved. Previously, based on simple geometric considerations, it was proposed that C6 diacridine could only span a single base pair, which requires that its bifunctional reaction violates the widely-observed "neighbor exclusion principle" where bound intercalators are separated by at least 2 base pairs. Here we have explored the structure of diacridine-DNA complexes using unrestrained molecular dynamics in explicit solvent using the parmbsc0 forcefield in AMBER14. We studied the C4 to C8 dimers, intercalated via both the minor and major DNA grooves, to a variety of nucleotide sequences. We find that C6, C7, and C8 diacridine are able to form 2 base pair bisintercalated complexes from either groove, whereas the C4 and C5 homologues cannot. We conclude that C6 diacridine does have the capacity to bisintercalate without violating neighbor exclusion, and that the previous proposed binding model needs revision.

Design and synthesis of a peptide derivative of ametantrone targeting the major groove of the d(GGCGCC)2palindromic sequence

We report the synthesis of a peptide derivative of antitumor anthraquinones, designed to target GC-rich palindromic sequences. It has micromolar activities on three cancer cell lines and is fifty times less toxic than mitoxantrone on a healthy line.

Drug-DNA intercalation: from discovery to the molecular mechanism

The ability of small molecules to perturb the natural structure and dynamics of nucleic acids is intriguing and has potential applications in cancer therapeutics. Intercalation is a special binding mode where the planar aromatic moiety of a small molecule is inserted between a pair of base pairs, causing structural changes in the DNA and leading to its functional arrest. Enormous progress has been made to understand the nature of the intercalation process since its idealistic conception five decades ago. However, the biological functions were detected even earlier. In this review, we focus mainly on the acridine and anthracycline types of drugs and provide a brief overview of the development in the field through various experimental methods that led to our present understanding of the subject. Subsequently, we discuss the molecular mechanism of the intercalation process, free-energy landscapes, and kinetics that was revealed recently through detailed and rigorous computational studies.

Stacking interactions and intercalative DNA binding

The DNA-binding properties of many ligands can be rationalized on the basis of their structural and electronic complementarity with the functional groups present in the minor and major grooves of particular DNA sequences. Specific hydrogen bonding patterns are particularly useful for the purpose of sequence recognition. Less obvious, however, is the influence of base composition on the conformational preferences of individual base steps and on the binding of intercalating moieties which become sandwiched between contiguous base pairs. Improved knowledge of stacking interactions may lead to a better understanding of the architecture and inherent flexibility of particular DNA sequences and may provide insight into the principles that dictate the structural changes and specificity patterns observed in the binding of some intercalating ligands to DNA.

Structure and energetics in the complexes of a double-stranded B-DNA dodecamer with netropsin derivatives of a tricationic water-soluble porphyrin: a theoretical investigation

The structural and energetical characteristics of the complexes formed between two auto-complementary DNA dodecamers, d(CGCGAATTCGCG)2, and d(GCGCAATTGCGC)2, and two novel netropsin (I) and glycine-netropsin (II) conjugates of a tricationic water-soluble porphyrin are investigated in detail by means of theoretical computations. This study was prompted by the successful chemical synthesis of II, which was recently reported (Anneheim-Herbelin, G., Perrée-Fauvet, M., Gaudemer, A., Hélissey, P., Giorgi-Renault, S. and Gresh, N., Tetrahedron Lett. 34, 7263 (1993)). The results indicate that: a) Intercalative binding of II does not entail significant distortions of the DNA backbone, and the Net moiety can bind tightly to the core of the minor groove. b) Intercalative binding of I is computed to energetically weaker than that of II. This is a consequence of the reduced length of the oligopeptide arm, such that the terminal propionamidinium group interacts less favorably with the fourth A-T base-pair than is the case with II. c) Nonintercalative binding of II produces considerable conformational distortions of the DNA. These results in a break of the DNA axis in between the fourth and the fifth base-pairs, namely, at the level where the long axis of the chromophore and of the oligopeptide intersect.

Binding site for ethidium cation in the major groove of B-form DNA

Molecular modeling of the system composed of ethidium cation initially intercalated through the major and minor grooves of the B-DNA form of 5'GCATGC3' into the central AT pairs, with the phenyl group oriented either toward or away from the helical axis, led upon energy minimization to intercalation complexes only when the phenyl group was initially oriented away from the helical axis. For the other two cases, energy minimization led to extrusion of the ethidium from the helix to form unanticipated outside-bound complexes stabilized by specific hydrogen bonds. It is suggested that outside-bound major groove complexes involving the GXXXG sequence may play important roles in the reaction mechanism of ethidium intercalation and in its action as a drug.

A theoretical investigation of the base sequence preferences of monointercalating polymethylene carboxamide derivatives 9-aminoacridine

Theoretical computations are performed of the comparative binding affinities of five polymethylene carboxamide derivatives of 9-aminoacridine to a series of double-stranded hexanucleotides. The purpose of this investigation is to ascertain whether minor groove recognition of a guanine base adjacent to the intercalation site can occur, and be preferentially stabilized, for a given length of the polymethylene side chain, encompassing from n = 2 up to n = 6 methylene groups. For that purpose, several representative sequences were investigated, in which intercalation of the 9-aminoacridine chromophore occurred at a central d(CpG) or d(TpA) step. Investigated were the self-complementary sequences d(CGCGCG)2, d(GCCGGC)2, d(TATATA)2 and d(ATTAAT)2, as well as the 'mixed' sequences d(ACTAAT) .d(ATTAGT) and d(TGTATA). d(TATACA). For n = 3 up to n = 6, such a recognition was enabled only when the guanine base was located downstream of the intercalation site, i.e. with steps d(CGG) and d(TAG). It occurred by means of a bidentate interaction involving, on the one hand, H(N2) and N3 of the base, and, on the other hand, the carbonyl oxygen and the cis amino hydrogen of the terminal formamide moiety of the ligand. Because of the flexibility of the side chain, however, alternative binding modes were also found to occur competitively, involving backbone-only interactions of the side chain. On the basis of the present computations, upon binding to the sequence d(GCCGGC)2, an optimal value of n = 5 could be derived, with the corresponding acridine derivative eliciting both a significant prevalence of the bidentate over backbone only binding mode, and the most favourable energy balance within the investigated series. This privileged value of n = 5 is fully consistent with the experimental results of Markovits et al. and Gaugain et al. The very flexibility of the side chain, however, hampered any preferential recognition of a triplet sequence with a downstream guanine, such as d(CGG) or d(TAG), to be elicited over sequences such as d(TAA), d(TAT) or d(TAC).

Strategies for the use of computational SAR methods in assessing genotoxicity

The relationship between computational SAR studies and relevant data gathering and generation activities is complex. First, the chemical class to be studied is selected on the basis of information requirements for hazard identification and assessment. Membership in the class is determined by consideration of chemical structure and reactivity. Compilation of the existing bioassay data for this chemical class follows immediately from the specification of the class. Bioassay data, qualitative knowledge of general chemical reactivities in this class, and knowledge concerning potential interactions with biomolecular targets all contribute to the derivation of possible mechanisms for biological activity. Computational studies based on modeling the proposed mechanism of action and/or the existing data base can provide a quantitative basis for the differentiation between chemicals. There is the opportunity for continuing feedback between the quantitative computational studies and the development of a relevant bioassay data base for this chemical class. The qualitative and quantitative information on the potential biological responses obtained will provide a rational basis for extrapolation from the extant data base to the chemicals of interest, and to biological responses significant to the assessment for which complete data are unavailable. Knowledge concerning possible mechanisms of action and preexisting data determine the type of computational study that will be most useful.

A theoretical study of anthracene and phenanthrene derivatives acting as A-T specific intercalators

Theoretical computations are performed on the comparative A-T versus G-C binding selectivities of two DNA intercalating molecules recently synthesized by Wilson et al. These are derivatives of phenanthrene and anthracene with side chains containing an hydroxy group bound to its C alpha carbon and a cationic amino group bound to its C beta carbon. We have optimized the binding energies of these phenanthrene and anthracene derivatives (1 and 2, respectively) to the double-stranded tetramers d(ATAT)2 and d(GCGC)2, the intercalation occurring in the central pyrimidine (3'-5') purine sequence. The sum of the intercalator-oligonucleotide intermolecular interaction energy plus the conformational energy variation of the intercalator upon binding were computed by the SIBFA procedures, which use empirical formulas based on ab initio SCF computations. Both compounds are found to bind more favourably to the AT sequence than to the GC one. Moreover, the affinity of 1 for the AT oligomer is computed to be larger than that of 2, whereas conversely that of 2 is larger than that of 1 for the GC oligomer. The AT versus GC binding selectivity of 1 is significantly larger than that of 2. These results are in excellent agreement with the experimental findings of Wilson et al. However, contrary to the suggestion of these authors the alpha-hydroxy group of the side chain of the intercalators does not seem to play a decisive role in determining the A-T specificity.

Computer modeling of actinomycin D interactions with double-helical DNA

We have performed molecular mechanical calculations on intercalation complexes of actinomycin D with a series of base-paired hexanucleoside pentaphosphates; d(GCGCGC)2, d(GCCGGC)2, d(GCATGC)2, d(GCTAGC)2 and d(ATGCAT)2. Our results are in good agreement with previous experimental work on sequence selectivity. The results provide a rationalization for the strong preference of actinomycin D to intercalate on the 3' side of guanine residues, consistent with previously proposed models. Finally, the computed structures for d(ATGCAT)2-actinomycin D complexes have been compared with two-dimensional nuclear magnetic resonance nuclear Overhauser effect experimental results. To our knowledge, this is the first extensive comparison of molecular mechanical model structures for a drug-DNA complex with experimental solution phase data. We find generally good agreement between our computational models and the experimental solution phase structures.

DNA sequence preferences for the anti-cancer drug mitoxanthrone and related anthraquinones revealed by DNase I footprinting

The interaction has been studied of several anthraquinone-based intercalating drugs, including the anti-cancer agent mitoxantrone, with defined sites of DNA. A 160 base pair DNA sequence from tyrT was employed for footprinting with DNase I. The anthraquinones had aminoalkylamino substituents in various positions of the ring system. Inhibition of enzymatic cutting of the DNA was observed at various positions on the sequence, mostly around some of the pyrimidine-3',5'-purine sites. Enhancements to cutting were observed clustered around AT-rich regions. The compounds showed differences in detailed footprinting behaviour, which have been related to differences in their mode of interaction with DNA as found in earlier computer modelling studies.

A theoretical investigation on the sequence selective binding of mitoxantrone to double-stranded tetranucleotides

Theoretical computations are performed on the comparative binding energetics of mitoxantrone (MX), a newly synthesized intercalating anthraquinone antitumor drug, to six representative double-stranded tetranucleotides: d(GCGC)2, d(CGCG)2, d(ATAT)2, d(TATA)2, d(GTGT), d(ACAC), and d(CCGG)2. The computations are performed with the SIBFA procedure, which uses empirical formulas based on ab initio SCF computations. The best binding configuration of mitoxantrone locates its two side chains in the major groove. A considerable preference is elicited for intercalation of the chromophore ring in a pyrimidine (3'-5') purine sequence rather than the isomeric purine (3'-5') pyrimidine sequence. Contrary to the situation encountered with "simple" intercalators, in which this preference is generally attributed solely to differences in the energies of unstacking necessary to generate the intercalation sites, the preference is dictated in MX to a large extent by the intermolecular interaction energy term. This result is imposed by the interactions of the side chains of MX with the oligonucleotide.

A theoretical investigation on the sequence selective binding of adriamycin to double-stranded polynucleotides

Theoretical computations are performed on the structural and energetical factors involved in the sequence selective binding of adriamycin (ADM) to five self-complementary double-stranded hexanucleotides. Among the two regularly alternating hexanucleotides d (TATATA)2 and d (CGCGCG)2, a stronger binding is predicted for the former. The strongest complex is computed, however, for the mixed hexanucleotide d (CGTACG)2, containing the intercalation site between two CG base pairs and an adjacent TA base pair. The overall sequence preference is the result of an intricate interplay of sequence preferences of the constituents in particular of daunosamine and the 9-OH substituent. Altogether, the selective base pair recognition by adriamycin cannot be defined in terms of the two base pairs implicated in the intercalation site alone but must be expressed in terms of a triplet of base pairs.

A theoretical investigation on the sequence selective binding of daunomycin to double-stranded polynucleotides

Theoretical computations are performed on the structural and energetical factors involved in the sequence selective binding of daunomycin (DNM) to six representative self-complementary double-stranded hexanucleotides: d(CGTACG)2,d(CGATCG)2,d(CITACI)2, d(TATATA)2, d(CGCGCG)2 and d(TACGTA)2. The conformational angles of the hexanucleotides are fixed in values found in the representative crystal structure of the d(CGTACG)2-DNM complex. The intermolecular DNM-hexanucleotide interaction energies and the conformational energy changes of DNM upon binding are computed and optimized in the framework of the SIBFA procedure, which uses empirical formulas based on ab initio SCF computations. Among the two regularly alternating hexanucleotides, d(TATATA)2 and d(CGCGCG)2, a stronger binding is predicted for the former, in agreement with experimental results obtained with poly(dA-dT).poly(dA-dT) and poly(dG-dC).poly(dG-dC). Altogether, however, among the six investigated sequences, the strongest complexes are computed for the mixed hexanucleotides d(CGATCG)2 and d(CGTACG)2, containing the intercalation site between two CG base pairs and an adjacent TA base pair. This situation may be related to the increased affinity of DNM for GC rich DNA's and to the situation in the crystal structure of the DNM-d(CGTACG)2 complex. Analysis of the intrinsic base sequence preferences expressed by the individual constituents of DNM, namely the daunosamine side chain, the chromophore ring and its two 9-hydroxyl and 9-acetoxy substituents, reveals that the overall sequence preference found is the result of a rather intricate interplay of intrinsic sequence preferences, in particular at the level of daunosamine and the 9-hydroxyl substituent.(ABSTRACT TRUNCATED AT 250 WORDS)

Structural and sequence-dependent aspects of drug intercalation into nucleic acids

Information gained from X-ray crystallographic studies on drug-nucleic acid complexes is described, with emphasis on the intercalation process. Relevant data from NMR experiments are examined in order to highlight similarities and differences between solution and solid-state structures. Theoretical analyses of intercalation complexes are also discussed and evaluated, with respect to the structural methods, with special reference being made to nucleic acid conformation and positions of drug molecules in the binding sites.

Selective inhibition of restriction endonuclease cleavage by DNA intercalators

The preferred dye binding sites and the microenvironment of known nucleotide sequences within mitochondrial and plasmid pBR322 DNA was probed in a gross fashion with restriction endonucleases. The intercalating dyes, ethidium bromide and propidium iodide, do not inhibit a given restriction endonuclease equally at all of the restriction sites within a DNA molecule. The selective inhibition may be explained, in part, by the potential B to Z conformation transition of DNA flanking the restriction site and by preferred dye binding sites. Propidium iodide was found to be a more potent inhibitor than ethidium bromide and the inhibition is independent of the type of cut made by the enzyme.

Binding of ethidium bromide to self-complementary deoxydinucleotides

Optical spectra titrations were performed with ethidium bromide and the self-complementary deoxydinucleotides pdCpdG, pdGpdC, pdTpdA, and pdApdT. The titrations were performed in 7.5 mM phosphate buffer, pH 7.0, at 7 degrees C, and with varying dinucleotide concentrations always in large excess of the dye concentration. Well-defined isosbestic points were present in each titration after correction for dinucleotide light absorption. The binding curves were evaluated in terms of simple bimolecular or termolecular reaction models. The bimolecular reaction model gave a significantly better fit to the experimental data, judging from a computerized nonlinear least-squares fitting procedure. The following equilibrium constants were obtained: KC-G = 2000 M-1; KG-C = 950 M-1; KT-A = 370 M-1; KA-T = 350 M-1. From these data the absorption spectra of the completely bound dye were evaluated. These spectra showed bathchromic shifts of their maxima, increasing with the magnitude of K. Fluorescence spectra of ethidium bromide/dinucleotide mixtures were recorded under conditions similar to those for absorption spectra. From the known equilibrium constants the contributions of the bound dye could be estimated. The following fluorescence enhancements Ib/If were found: IC-Gb/If = 6.5; IG-Cb/If = 3.0; IT-Ab/If = 2.0; IA-bT/If = 2.0. From our results, in relation to other theoretical and experimental studies, we conclude that electrostatic phosphate-dye interactions give rise to a major part of the binding energy, which varies with dinucleotide geometry. The more strongly bound complexes exhibit less exposure of the dye to the solvent.

Quantum chemical calculations on the two-step mechanism of proflavin binding to DNA

Quantum chemical calculations on the binding of proflavin to DNA lead to a model in which the outside binding to a phosphate group leads to an induced fit in the intercalation receptor site. The calculations suggest hydrogen bonding of the amine groups of the outside bound proflavin to the anionic oxygen of the backbone phosphate. The resulting partial neutralization facilitates the conformational transitions required for intercalation. The results are consistent with the observed preference of proflavin for dCpdG over dGpdC sequences and with the observed kinetics of the binding reaction.

Conformational flexibility of dinucleoside dimers during unwinding from the B-form to an intercalation structure

Intra-dinucleoside dimer conformational energy calculations were performed on a series of conformations corresponding to the unwinding of a dinucleoside dimer from the B-form type structure. Conformational states were characterized in terms of the distance, d, between parallel bases. All 16 possible sequences for (G) and (C) and/or (A) and (T) bases were considered. Both free space, and free space plus aqueous solvation energies were computed. It was found that the B-form like conformer is the most stable structures in free space and in an aqueous medium. However, the energy as function of d is base pair sequence dependent. This sequence dependence suggests a possible inherent specificity for intercalation of the dinucleoside dimer with a drug molecule.

Binding of platinum(II) intercalation reagents to deoxyribnonucleic acid. Dependence on base-pair composition, nature of the intercalator, and ionic strength

The DNA binding of three platinum(II) intercalation reagents has been studied and found to depend upon base composition, the nature of the intercalator, and the ionic strength of the solvent medium. In 0.2 M NaCl, binding data for calf thymus DNA show the association constants to be approximately 10(4) M-1. The binding constants decrease in the order [(o-phen)Pt(en)]2+ greater than or equal to [(terpy)Pt(HET)]+ greater than [(bipy)Pt(en)]2+. The number of available intercalation sites for the doubly charged intercalators is only 70% of the number expected from the nearest-neighbor exclusion model. Binding of [(o-phen)Pt(en)]2+ and [(terpy)Pt(HET)]+ to various DNAs depends linearly on G.C content. Both reagents exhibit essentially the same degree of G.C specificity. Intercalative binding is a function of ionic strength. Increasing the salt concentration minimizes the importance of metallointercalator charge, and extrapolation to 1 M salt reveals the intercalative abilities, as reflected in binding constants, to be equivalent for [(terpy)Pt(HET)]+ and [o-phen)Pt(en)]2+ and about 1 order of magnitude less than that of ethidium.