Interactions of molecules with nucleic acids. XI. Generalization of techniques to generate nucleic acid structures with applications to intercalation sites and kinked structures

A generalized procedure to generate nucleic acid structures is presented. In this procedure, the bases of a base pair are oriented first for characterization of particular DNA receptor sites. The resultant sites are then used in the study of specific molecule–DNA interactions. For example, intercalation sites, kinked DNA, and twisted and tilted bases are envisioned. Alterations of structures via anti → syn orientations of bases, as well as crankshaft motion about collinear bonds, provide additional conformations without disrupting the overall backbone structure. These approaches to the generation of nucleic acid structures are envisioned as required in studies of the intercalation phenomenon, minor adjustments of DNA to accommodate denaturation, binding of carcinogens to DNA, complex formation of transition metals with DNA, and antitumor agents as ligands. For these base-pair and base orientations, backbone orientations are calculated by the AGNAS technique to yield physically meaningful conformations, namely, those conformations for which nonbonded contacts are favourable. A procedure is presented to generate dimer duplex units that are physically meaningful and to assemble these units into a polynucleotide duplex. Double helices that begin with B-DNA, undergo a transition to one of the above-mentioned receptor sites, and return to B-DNA can be assembled from a catalog of dimer duplexes. Stereographic projections of the various receptor sites already being used to model binding to DNA are presented.     

Interactions of molecules with nucleic acids. IV. Binding energies and conformations of acridine and phenanthridine compounds in the two principal and in several unconstrained dimer-duplex intercalation sites

The binding positions and relative minimum binding energies are calculated for complexes of 9-aminoacridine, proflavine, N-methylphenanthridinium, and ethidium in theoretically determined intercalation sites in B-DNA (sites I and II) and in unconstrained dimer-duplex sites. The selection of site I in B-DNA by these compounds agrees with the theoretical interpretation of studies of unwinding angles in closed circular DNA in all cases but ethidium, which is predicted to select site II. The most stable binding positions of the acridines and ethidium in unconstrained dimer-duplex units agree with experimental results of intercalation complexes of dinucleoside monophosphate units. Base-pair specificity for Watson-Crick pairing is examined. The energy of an intercalation complex is partitioned into ΔE₂₃, the energy required to open base pairs BP₂ and BP₃ in B-DNA to a site, and ΔE_In, the energy change when a free molecular intercalates. ΔE₂₃ depends strongly on the base-pair sequence, whereas ΔE_In for the four molecules studied does not. The three most stable sequences contain (pyrimidine)p(purine) units, and this provides a rationale for the exclusive formation of crystals of intercalation complexes with these units. In spite of this selectivity, the distribution of G?C and A?T base pairs is equal for these three units and persists as the more unstable sequences are included. Therefore, specificity arises from the interaction between the base pairs and the 2′-deoxyribose 5′-monophosphate backbone for the opening of B-DNA to an intercalation site and not from the interaction between the chromophore and the DNA.     

Interaction of molecules with nucleic acids. II. Two pairs of families of intercalation sites, unwinding angles, and the neighbor-exclusion principle

Intercalation-site geometries are generated for a tetramer duplex extracted from B-DNA. Glycosidic angles and puckers of the deoxyribose sugar groups bonded to base pairs BP₁ and BP₄, namely, those at either end of the tetramer duplex, are assumed to be those of B-DNA to insure continuity. All possible geometrical conformations for combinations of C(2′)-endo, C(3′)-endo, C(2′)-exo, and C(3′)-exo sugar puckers are determined for the tetranucleotide backbone. Those with minimum energy are selected as candidates for intercalation sites. Calculations reveal two pairs of physically meaningful families of intercalation sites which occur in two distinct regions, I and II, of helical angles which orient BP₂ relative to BP₃ and with the helical axis disjointed between these base pairs. For each site I and II within BP₂ and BP₃, there are two distinct backbone conformations, A and B, connecting BP₃ to BP₄ or BP₁ to BP₂ which do not disrupt backbone conformations connecting BP₂ to BP₃. Hence two pairs, IA and IB, and IIA and IIB, of intercalation sites exist in which the sugar puckers along the backbone of the tetramer alternate from C(2′)-endo to C(3′)-endo on the backbone (5′p3′) connecting BP₂ to BP₃. The glycosidic angles of the C(3′)-endo sugar χ₃^γ are, coincidentally, 80° ± 2° for both conformations γ = A and B connecting BP₃ to BP₄ along the phosphate backbone (5′p3′). Consistent with the theoretical results, the experimental unwinding angles can be grouped into two categories with absolute values of 18° and 26°. The theoretical unwinding angles for sites IA and IB of 16° and for sites IIA and IIB of 20° occur for a displacement of -0.8 Å in the helical axes of BP₂ and BP₃ and for a 100% G·C composition, with a decrease depending on the amount of A·T base pairs present. Ratios of theoretical unwinding angles of sites I and II, which range from 0.75 to 0.84 for the two principal sites, compare well with the experimental value of 0.71. The theoretical results, in agreement with experimental observation, provide a new interpretation of the nature and conformation of the possible binding sites. Conformations obtained from these studies of intercalation sites in a tetramer duplex are used to rationalize the well-known neighbor-exclusion principle. The possibility of violation of this principle is demonstrated by the existence of two families of physically meaningful conformations. Conformations of unconstrained dimer duplexes are also obtained, one of which corresponds to the experimental crystal structure of ethidium–dinucleoside complexes, but these cannot be joined to the B-DNA structure. Backbone conformations of the tetramer duplex can be constructed until the base-pair separation reaches 8.25 Å, which may limit the molecules that can intercalate.     

Interactions of molecules with nucleic acids. VI. Computer design of chromophoric intercalating agents

The mode of action of many antitumor agents entails the inhibition of nucleic acid synthesis. Because many of the drugs can intercalate, it is assumed that intercalation is an important step in the mechanism of biological activity. As intercalants contain a planar chromophore as an ingredient essential for intercalation, chromophores that should fit into DNA are desired. This is the main theme of this investigation. Binding to DNA of fundamental moieties, protonated pyridine, aniline, phenol, quinone, and 4H-thiopyran-4-one, is studied to determine their optimum placement in DNA. The optimum orientations for each moiety are superimposed to form polyaromatic systems that can intercalate in a manner in which functional groups on these chromophores are oriented as in the moieties themselves. Ideal intercalants proposed contain three and four fused ring system, have protonated ring nitrogen atoms located to maximize the electrostatic interactions with DNA, hydroxy and amino groups that can hydrogen bond to the O_II and O_5′ phosphate backbone atoms, and carbonyl and sulfur groups in the central position of the ring system to provide variations in the chromophore and to interact with the relatively positive region in the intercalation site. The optimum orientation occurs when the chromophore and the base pairs overlap to the maximum extent. The ideal intercalants are fundamentally of the type:      

Ethidium bromide-dinucleotide complexes. Evidence for intercalation and sequence preferences in binding to double-stranded nucleic acids.

The solution complexes of ethidium bromide with nine different deoxydinucleotides and the four self-complementary ribodinucleoside monophosphates as well as mixtures of complementary and noncomplementary deoxydinucleotides were studied as models for the binding of the drug to DNA and RNA. Ethidium bromide forms the strongest complexes with pdC-dG and CpG and shows a definite preference for interaction with pyrimidine–purine sequence isomers. Cooperativity is observed in the binding curves of the self-complementary deoxydinucleotides pdC-dG and pdG-dC as well as the ribodinucleoside monophosphates CpG and GpC, indicating the formation of a minihelix around ethidium bromide. The role of complementarity of the nucleotide bases was evident in the visible and circular dichroism spectra of mixtures of complementary and noncomplementary dinucleotides. Nuclear magnetic resonance measurements on an ethidium bromide complex with CpG provided evidence for the intercalation model for the binding of ethidium bromide to double-stranded nucleic acids. The results also suggest that ethidium bromide may bind to various sequences on DNA and RNA with significantly different binding constants.     

Interactions of molecules with nucleic acids. XII. Theoretical model for the interaction of a fragment of bleomycin with DNA

An intercalation model of a complex between DNA and a bleomycin fragment (BLMF), consisting of the bithiazole core and an amide and a protonated amino substituent, is presented. The model, which shows a preference for BLMF with the protonated amine in the minor groove and the acetyl terminal inserted into either the minor and major grooves, respectively, agrees with recently obtained nmr data. The selection of sites I and II, which have the smallest unwinding of the three theoretical intercalation sites, is consistent with the experimental unwinding angle of 12°. The bithiazole moiety stacks between two base pairs of the double helix, while the protonated substituent interacts ionically with the negatively charged regions of the backbone in the minor groove of the DNA. The protonated amine also forms an intramolecular hydrogen bond with the carbonyl oxygen of the amide group on the same substituent. Analysis of drug complexes with different base-pair sequences reveal four energetically defined groups. The relative energy of the dimer duplex complexes of BLMF correlates with bleomycin's observed base-sequence specificity upon cleavage. The most stable intercalation complexes form adjacent to the bases cleaved most readily. This correlation suggests a primary connection between intercalation and cleavage. A model cleavage site based on these preliminary theoretical calculations and the experimental observations is proposed. It consists of an intercalation site in a trimer duplex. Pyrimidine(p)purine sequences are the predominant sites for intercalation, and the base adjacent to the site at the (3′) end is cleaved.     

The binding of metabolites formed from aminostilbene derivatives to nucleic acids in the liver of rats.

Carcinogenic trans-4-dimethylaminostilbene (trans-DAS) and trans-4-acetylaminostilbene (trans-AAS) as well as inactive cis-DAS and DABB were highly and specifically labeled with tritium and administered orally to female Wistar rats. Covalent binding to liver rRNA and DNA was measured and found to be higher for the carcinogenic compounds. Digests from these nucleic acids were chromatographed on Sephadex LH-20 and 16 different nucleoside adducts were characterised by their retention volumes. Labeled trans-DAS was administered in doses ranging from 0.025--250 mumol/kg. Binding to nucleic acids was directly proportional to the dose at low doses (0.025--2.5 mumol/kg) and less than proportional at higher doses (25--250 mumol/kg). The pattern of nucleoside adducts remained practically constant over the wide range of doses. A pharmacokinetically determined threshold of metabolic activation thus could not be demonstrated for this compound. A modified procedure is described to simultaneously isolate pure liver rRNA and DNA from nonfasted rats in high yields.     

Interactions of molecules with nucleic acids. VII. Intercalation and T.A specificity of daunomycin in DNA

Intercalation complexes of daunomicin(+1) with tetramer duplexes in DNA are studied with the theoretically determined intercalation sites (I, ?0.4), (II, ?0.4), and (III, ?1.4). These sites occur with base pairs separated by 6.76 Å for helical angles of 26°, 22°, and 8° about the intercalation site. Site I is preferred, and this is in agreement with experimental unwinding angles. Optimum binding positions and conformations are established, and these are in agreement with experimental results from crystal structures. A systematic procedure is devised to study base-pair and base-sequence specificity, which results in the demonstration that the most stable sequences are mainly ↑BP₁, T·A, DAUN, A·T, BP₄↓ and ↑BP₁, T·A, DAUN, G·C, BP₄↓, i.e., with the TpA and CpG (pyrimidine)p(purine) sequences about the intercalation site. These 32 possible sequences are found among the 40 most stable complexes. These theoretical calculations of intercalation complexes with daunomicin(+1) provide the first example in which a drug specifically selects the base pair T·A and prefers it in a particular sequence about the intercalation site. This specificity is in agreement with some experimental results. Problems associated with the interpretation of specificity are discussed in terms of the base, base-pair, and base-sequence resulting from the DNA site and the DNA–drug interactions. T·A specificity is rationalized by noting that the 2′deoxyribo-5′-monophosphate backbone attached to A is slightly more negative than that on the other nucleotides. Hence, a preference exists for binding to the protonated daunosamine (+1) groups. Stereographic projections of daunomycinone and daunomycin(+1) in a bond model and in a space-filling model with steric contours illustrate the results.     

Interactions of molecules with nucleic acids. I. An algorithm to generate nucleic acid structures with an application to the B-DNA structure and a counterclockwise helix.

An algorithm is developed that enables the routine determination of backbone conformations of nucleic acids. All atomic positions including hydrogen are specified in accord with experimental bond lengths and angles but with theoretically determined conformational angles. For two Watson-Crick base pairs at a separation of 3.38 Å, and perpendicular to a common helical axis, minimum energy configurations are found for all 10 combinations at helical angles of α ～ 36°–38°, corresponding to the B-DNA structure with C(2′)-endo sugar puckers. Backbone configurations exist only within the range 35.5° ? α ? 42°, which suggests the origin of the 10-fold helix. Calculated stacking energies for the B-DNA structure increases for each of the clustered groups of base pairs: G·C with G·C, G·C with A·T, and A·T with A·T, and they are in approximate agreement with experimental observations. The counter-clockwise helix is examined, and physically meaningful structures are found only when the helical axes of successive base pairs are disjointed.     

Determination of interactions between structured nucleic acids by fluorescence resonance energy transfer (FRET): selection of target sites for functional nucleic acids.

We previously developed a method for monitoring the integrity of oligonucleotides in vitro and in vivo by quantitating fluorescence resonance energy transfer (FRET) between two different fluorochromes attached to a single oligonucleotide. As an extension of this analysis, we examined changes in the extent of FRET in the presence or absence of target nucleic acids with a specific sequence and a higher-ordered structure. In this system FRET was maximal when probes were free in solution and a decrease in FRET was evidence of successful hybridization. We used a single-stranded oligodeoxyribonucleotide labeled at its 5'-end and its 3'-end with 6-carboxyfluorescein and 6-carboxytetramethylrhodamine, respectively. Incubation of the probe with a single-stranded complementary oligonucleotide reduced the FRET. Moreover, a small change in FRET was also observed when the probe was incubated with an oligonucleotide in which the target site had been embedded in a stable hairpin structure. The decrease in the extent of FRET depended on the length of the stem region of the hairpin structure and also on the higher-ordered structure of the probe. These results indicate that this spectrofluorometric method and FRET probes can be used to estimate the efficacy of hybridization between a probe and its target site within highly ordered structures. This conclusion based on changes in FRET was confirmed by gel-shift assays.     

Interactions of bacteriophage T4-coded gene 32 protein with nucleic acids: I. Characterization of the binding interactions

In this paper we examine molecular details of the interaction of bacteriophage T4-coded gene 32 protein with oligo- and polynucleotides. It is shown that the binding affinity (K_oligo) of oligonucleotides of length (l) from two to eight nucleotide residues for gene 32 protein is essentially independent of base composition or sugar type. This binding also shows little dependence on salt concentration and on oligonucleotide length; even the expected statistical length factor in K_oligo is not observed, suggesting that binding occurs at the end of the oligonucleotide lattice and that the oligonucleotide is not free to move across the binding site. Co-operative (contiguous) or isolated binding of gene 32 protein to polynucleotides is very different; here binding is highly salt dependent ($\text{? log Kω}\text{? log}\text{[NaCl]}\text{～- ?7}\text{)}$ and essentially stoichiometric at salt concentrations less than ~0.2 m (for poly(rA)). Binding becomes much weaker and the binding isotherms appear typically co-operative (sigmoid) in protein concentration at higher salt concentrations. We demonstrate, by fitting the co-operative binding isotherms to theoretical plots at various salt concentrations and also by measuring binding at very low protein binding density (ν), that the entire salt dependence of Kω is in the intrinsic binding constant (K); the co-operativity parameter (ω) is essentially independent of salt concentration. Furthermore, by determining titration curves in the presence of salts containing a series of different anions and cations, it is shown that the major part of the salt dependence of the gene 32 protein-polynucleotide interaction is due to anion (rather than to cation) displacement effects. Binding parameters of oligonucleotides of length sufficient to bind two or more gene 32 protein monomers show behavior intermediate between the oligonucleotide and the polynucleotide binding modes. These different binding modes probably reflect different conformations of the protein; the results are analyzed to produce a preliminary molecular model of the interactions of gene 32 protein with nucleic acids in its different binding modes.     

Interactions of the dimethyldiazaperopyrenium dication with nucleic acids. 2. Binding to double-stranded polynucleotides

The interactions of dimethyldiazaperopyrenium dication (1) with DNA have been studied by spectroscopic methods: absorption, static and dynamic fluorescence, and linear dichroism. 1 binds strongly to DNA at 250 mM NaCl, with a higher affinity for G-C pairs as compared to A-T pairs. The dye fluorescence is enhanced when it is bound to A-T pairs, whereas the emission is quenched in the vicinity of G-C pairs. Evidence for intercalation has been obtained via energy transfer and linear dichroism measurements.     

Assessment of methods for covalent binding of nucleic acids to magnetic beads, Dynabeads, and the characteristics of the bound nucleic acids in hybridization reactions.

Dynabeads are magnetic monosized beads with high stability, high uniformity, unique paramagnetic properties, low particle-particle interaction, and high dispersibility. Different reactive groups; hydroxyl, carboxyl and amino groups can be attached to the surface. Several methods for covalent attachment of DNA or oligonucleotides to the beads were investigated. Best coupling yields were obtained by carbodiimide-mediated end-attachment of 5'-phosphate and 5'-NH2 modified nucleic acids to respectively amino and carboxyl beads. The carboxyl beads showed a low degree of non-specific binding, while a better yield of end-attached nucleic acids was obtained using the amino beads. The DNA-beads worked efficiently in hybridization experiments, and the kinetics of hybridization approach those of solution hybridization.     

The non-convalent binding of small molecules by ligandin. Interactions with steroids and their conjugates, fatty acids, bromosulphophthalein carcinogens, glutathione and realted compounds.

1. Equilibrium dialysis studies have been made of the binding of a number of small molecules by rat ligandin. Direct measurements of binding together with competition experiments indicated that bromosulphophthalein, oestrone sulphate and dehydroepiandrosterone sulphate each bind at the same single primary binding site with association constants of 1.1 X 10(7), 6.6 X 10(5) and 2.6 X 10(5) 1/mol respectively at pH 7.0,IO.16M,4 degrees C. As well as bromosulphophthalein and dehydroepiandrosterone sulphate, a number of strucurally similar organic anions including 2-hydroxyoestradiol-glutathione oestrone glycyronide, N-methyl-4-aminoazobenzene-glutathione and several bile acids, were able to displace oestrone sulphate from ligandin in a manner consistent with competition at a single binding site. From these experiments association constants for the competing ligands were derived; these were inthe range 1 X 10(4)-1 X 10(6) 1/mol. 2. Ligandin was found to bind a number of compounds for which, because of their low aqueous solubilities relative to their binding affinities complete binding isotherms could bot be obtained. These included several steroids (but not cortisol), 20-methylcholanthrene, diethylstilboestrol, oleate and palmitate. Oestrone sulphate was able to compete with these ligands for binding and the results of the competition experiments were interpretable in terms of 1:1 competition at a single binding site. 3. In general the conjugation of non-polar ligands with sulphate or glutathione resulted in increased affinities, but such increases were relatively small (approximately 15% in therms of free energy) implying that the main driving force for the binding of both the conjugated and unconjugated species was the hydrophobic effect. This conclusion is borne out by the observations that both oestrone and its sulphate showed slight increases in affinity with increase in ionic strength, as would be expected for hydrophobic interactions. 4. As well as non-polar compounds and organic anions, ligandin was also found to bind sulphate and glucuronate to a measurable degree, and to interact quite strongly with glutathione. For the latter compound a single binding site was found with an association constant of 1 X 10(5) 1/mol. Glutathione was able to cause the dissociation of the ligandin-oestrone sulphate complex, but this effect was not explicable in terms of simple 1:1 competition. 5. Both oestrone and oestrone sulphare were bound most strongly at pH 6-7, the affinity of the protein for these ligands falling off quite sharply on either side of this maximum. 6. The affinities of ligandin for bromosulphophthalein, steroids and their conjugates, diethylstilboestrol and N,N-dimethyl-4-aminoazobenzene are similar in magnitude to those of serum albumin and aminoazodye-binding protein A (B. Ketterer, E. Tipping, J.F. Hackney and D. Beale, 1976).     

Interactions of acridine orange with double stranded nucleic acids. Spectral and affinity studies

Spectral properties of acridine orange (AO) alone or in complexes with natural and synthetic nucleic acids of various base composition have been studied in aqueous solutions by absorption and fluorescence spectroscopy. The dimerization constant and absorption spectra of the dye in monomeric and dimeric form were established; dimerization of AO resulted in quenching of its fluorescence. Complexes of the dye with synthetic nucleic acids differed in the degree of enhancement of fluorescence quantum yield, varying between 1.42 to 2.38 fold as compared to AO monomer; these differences, however, were not base-dependent. Affinity of the dye to natural and synthetic polymers was studied and analyzed using McGhee-von Hippel model of polymer-ligand interactions. Because the sterical requirement for intercalative binding assumes interaction of dye monomer, the correction for AO dimerization was made in all calculations. All studied DNAs (natural and synthetic ones, the latter being homopolymer pairs or alternating copolymers of A,T or G,C or I,C base composition) had similar intrinsic association constants (KI = 5 X 10(4) - 1 X 10(5), M-1) and binding site size (n = 2.0-2.4 b.p.). The exception was poly(dA).poly(dT), having KI = 1.2 X 10(4) and n = 19.3 b.p. The results of KI measurement for calf thymus DNA and AO in different sodium ion concentration were in good agreement with predictions of the counterion condensation theory. The intercalation of AO into DNA is discussed in view of recent theoretical models of DNA-ligand interactions.     

Prediction of the binding energy for small molecules,peptides and proteins

A fast and reliable evaluation of the binding energy from a single conformation of a molecular complex is an important practical task. Knowledge‐based scoring schemes may not be sufficiently general and transferable, while molecular dynamics or Monte Carlo calculations with explicit solvent are too computationally expensive for many applications. Recently, several empirical schemes using finite difference Poisson–Boltzmann electrostatics to predict energies for particular types of complexes were proposed. Here, an improved empirical binding energy function has been derived and validated on three different types of complexes: protein–small ligand, protein–peptide and protein–protein. The function uses the boundary element algorithm to evaluate the electrostatic solvation energy. We show that a single set of parameters can predict the relative binding energies of the heterogeneous validation set of complexes with 2.5 kcal/mol accuracy. We also demonstrate that global optimization of the ligand and of the flexible side‐chains of the receptor improves the accuracy of the evaluation. Copyright © 1999 John Wiley & Sons, Ltd.