Modulation of acridine mutagen ICR191 intercalation to DNA by methylxanthines—Analysis with mathematical models

Caffeine (CAF) and other methylxanthines (MTX) may interact directly with several aromatic, intercalating ligands through mixed stacking aggregation. Formation of such stacking hetero-complexes may decrease their free form concentration and, in consequence, diminish their biological activity, which is often related to their direct interaction with DNA. In this paper interactions of acridine mutagen (ICR191) with DNA in the presence of three MTX: caffeine (CAF), pentoxifylline (PTX) and theophylline (TH) are investigated. Several mathematical models are used to calculate all association constant values and every component concentration in each analyzed mixture. Model McGhee–von Hippel is used to analyze ligand–DNA interaction, and model Zdunek et al.—to analyze ligand–MTX interactions. Finally, two distinct mathematical models are employed to analyze three-component mixture containing ligand, MTX and DNA molecules. The first model describes possible interactions of ligand with DNA and MTX, and rejects direct MTX interactions with DNA. The second model describes all interactions mentioned above and, additionally, allows MTX to interact directly with DNA. Results obtained using these models are similar. However, correspondence of theoretical results to experimental data is better for the first model than the second one. In this paper possible interactions of ICR191 with eukaryotic cell chromatin are also analyzed, showing that CAF reduces acridine mutagen potential to interact directly with cell chromatin. Additionally, it is demonstrated that MTX inhibit mutagenic activity of ICR191 in a dose-dependent manner. Furthermore, biological activity of ICR191–MTX mixtures corresponds with concentration of free mutagen form calculated using appropriate mathematical model.     

Thermodynamical model of mixed aggregation of ligands with caffeine in aqueous solution. Part II

A statistical-thermodynamical model of mixed association in which one component's self-association is unlimited while the second component does not self-aggregate is described. The model was tested with 4',6-diamidino-2-phenyl-indole-dihydrochloride (DAPI) and ethidium bromide (EB) using light absorption spectroscopy and calorimetry. The system is controlled by two parameters, which represent self-aggregation 'neighborhood' association constant KCC and mixed 'neighborhood' association constant KAC. Calculated, using this model, KAC = 58.2 +/- 1 M-1, KAC = 64.6 +/- 2 M-1 for DAPI and EB, respectively, are in good agreement with known values of stacking interactions. The titration microcalorimetric measurement of DAPI-CAF interaction delta H = -11.1 +/- 0.4 kcal/mol is also consistent with this type of reaction. The structures of the stacking complexes were also confirmed by semi-empirical molecular modeling in the presence of water. The data indicate that CAF forms stacking complexes with DAPI and EB, thus effectively lowering the concentration of the free ligands in the solution, and therefore, CAF can be used to modulate aromatic compound activity.     

Hetero-association of caffeine and aromatic drugs and their competitive binding with a DNA oligomer

NMR spectroscopy has been used to elucidate the molecular basis of the action of caffeine (CAF) on the complexation with DNA of mutagens such as ethidium bromide, propidium iodide, proflavine and acridine orange, and anticancer drugs such as actinomycin D and daunomycin. The hetero-association of CAF and each of the aromatic ligands in 0.1 mol L(-1) phosphate buffer (pD=7.1) has been investigated as a function of concentration and temperature by 500 MHz 1H NMR spectroscopy and analysed in terms of a statistical-thermodynamic model, in which molecules form indefinite aggregates for both self-association and hetero-association. The analysis leads to determination of the equilibrium constants of hetero-association and to the values of the limiting chemical shifts of the heteroassociation of CAF with each of the aromatic molecules. The hetero-association constants between CAF and each of the aromatic drugs/dyes are found to be intermediate in magnitude between those for self-association of CAF and the corresponding drug/dye. The most probable structures of the 1:1 CAF + ligand hetero-association complexes have been determined from the calculated values of the induced limiting chemical shifts of the drug protons. Knowledge of the equilibrium constants for self-association of CAF and the aromatic ligands, for their hetero-association and their complexation with a DNA fragment, the deoxytetranucleotide 5'-d(TpGpCpA), enabled the relative content of each of the CAF-ligand and CAF-ligand-d(TGCA) complexes to be calculated as a function of CAF concentration in mixed solutions. It is concluded that, on addition of CAF to the solution, the decrease in binding of drug or mutagen with DNA is due both to competition for the binding sites by CAF and the aromatic molecules, and to formation of CAF-ligand hetero-association complexes in the mixed solution; the relative importance of each process depends on the drug or mutagen being considered.     

Mixed mode of ligand-DNA binding results in S-shaped binding curves

S-shaped binding curves often characterize interactions of ligands with nucleic acid molecules as analyzed by different physico-chemical and biophysical techniques. S-shaped experimental binding curves are usually interpreted as indicative of the positive cooperative interactions between the bound ligand molecules. This paper demonstrates that S-shaped binding curves may occur as a result of the "mixed mode" of DNA binding by the same ligand molecule. Mixed mode of the ligand-DNA binding can occur, for example, due to 1) isomerization or dimerization of the ligands in solution or on the DNA lattice, 2) their ability to intercalate the DNA and to bind it within the minor groove in different orientations. DNA-ligand complexes are characterized by the length of the ligand binding site on the DNA lattice (so-called "multiple-contact" model). We show here that if two or more complexes with different lengths of the ligand binding sites could be produced by the same ligand, the dependence of the concentration of the complex with the shorter length of binding site on the total concentration of ligand should be S-shaped. Our theoretical model is confirmed by comparison of the calculated and experimental CD binding curves for bis-netropsin binding to poly(dA-dT) poly(dA-dT). Bis-netropsin forms two types of DNA complexes due to its ability to interact with the DNA as monomers and trimers. Experimental S-shaped bis-netropsin-DNA binding curve is shown to be in good correlation with those calculated on the basis of our theoretical model. The present work provides new insight into the analysis of ligand-DNA binding curves.     

Relaxation kinetics of DNA-ligand binding including direct transfer

The relaxation kinetics of binding of ethidium to calf-thymus DNA as studied previously by the temperature-jump method with absorption detection [Bresloff, J. & Crothers, D. M. (1975) J. Mol. Biol. 95 , 103] is reanalyzed in terms of a series of models for DNA-ligand interactions that include cases with and without internal and bimolecular direct transfer of ligands between different binding modes or different binding sites. The experimental results are shown to be consistent with two alternative binding mechanisms. Both models include bimolecular “direct transfer” steps and a site size of two base pairs per site. In the first model, there exist two distinct modes of binding to the DNA double helix at each binding site. In the second model, sites containing at least one GC base pair constitute a different and stronger class of binding sites than those containing only AT base pairs. The existence of a direct transfer pathway between two classes of bound ligands is supported by the linear increase of reciprocal relaxation times far beyond the concentration regions, where off-rates should have become rate limiting. The second model, incorporating the notion of base selectivity, is in quantitative agreement with published work on ethidium binding to DNAs of different base compositions and with nmr measurements of bound ligand lifetimes.     

Calculation of DNA condensation, caused by adsorption of ligands

A method for calculating the curves of DNA transition from linear to condensed state upon binding of condensing ligands has been developed. The character of the transition and ligand concentration necessary for condensation have been shown to be governed by the length of DNA molecule, energy and stoichiometry parameters of DNA-ligand complex (equilibrium constant between linear and condensed form in the absence of ligands, constants for ligand binding to linear and condensed forms, the number of base pairs covered by one ligand, etc.). The results of the calculations indicate that only slight difference in the free energies of these states in free DNA (less than 6 cal/mole(bp) for DNA of 500 bp long) is sufficient for the existence of stable linear state in the absence of ligands (in free DNA) and the formation of stable condensed state upon complexation.     

Binding free energy of selected anticancer compounds to DNA--theoretical calculations

Many studies have elucidated structures and thermodynamics of complexes formed by different ligands with DNA. However, in most cases structural and free energy binding studies were not correlated with each other because of the problem of identifying which experimental free energy of binding corresponds to which experimental DNA-ligand structure. In the present work, Poisson-Boltzmann and solvent-accessible surface area methods were used to predict unknown modes of interaction between DNA and three different ligands: mitoxantrone and two pyrimidoacridine derivatives. In parallel, experimental measurements of binding free energy for the studied complexes were performed to compare experimental and calculated values. Our studies showed that the calculated values of free energy are only close to experimental data for some models of interaction between ligands and DNA. Based on this correlation, the most likely models of DNA-ligand complexes were postulated: (i) mitoxantrone and one derivative of pyrimidoacridine, both with two charged side chains, intercalate from the minor groove of DNA and bind with both chains in this groove; (ii) pyrimidoacridine, with only one side chain, very likely does not intercalate into DNA at all. Additionally, the non-electrostatic and electrostatic parts of the calculated binding free energy for the DNA-ligands studied are discussed.     

Study of effect of water on the interaction of DNA and actinocin derivatives with various aminoalkyl chain length by infrared spectrophotometry and computer modeling

A comparative study of the effect of water on the interaction of DNA with actinocin derivatives having different numbers of methylene groups in side chains was performed by IR spectroscopy. It was found that, as relative humidity increases, water molecules simultaneously bind to hydrate-active sites of DNA and ligands. The absorption band at v = 1137 cm-1, caused by oscillations of the C-O and P-O groups of atoms in the DNA-ligand complex having two methylene groups, is due to the interactions between the cationic groups of the ligand and the sugar-phosphate backbone of DNA, which may be one of the reasons for the high stability of this complex. Using computer simulation of interaction of DNA fragments and actinocin derivatives in water environment, molecular models of the formation of their complexes for two ways of binding were constructed.     

A ruthenium dimer complex with a flexible linker slowly threads between DNA bases in two distinct steps

Several multi-component DNA intercalating small molecules have been designed around ruthenium-based intercalating monomers to optimize DNA binding properties for therapeutic use. Here we probe the DNA binding ligand [μ-C4(cpdppz)₂(phen)₄Ru₂]⁴⁺, which consists of two Ru(phen)₂dppz²⁺ moieties joined by a flexible linker. To quantify ligand binding, double-stranded DNA is stretched with optical tweezers and exposed to ligand under constant applied force. In contrast to other bis-intercalators, we find that ligand association is described by a two-step process, which consists of fast bimolecular intercalation of the first dppz moiety followed by ∼10-fold slower intercalation of the second dppz moiety. The second step is rate-limited by the requirement for a DNA-ligand conformational change that allows the flexible linker to pass through the DNA duplex. Based on our measured force-dependent binding rates and ligand-induced DNA elongation measurements, we are able to map out the energy landscape and structural dynamics for both ligand binding steps. In addition, we find that at zero force the overall binding process involves fast association (∼10 s), slow dissociation (∼300 s), and very high affinity (K_d ∼10 nM). The methodology developed in this work will be useful for studying the mechanism of DNA binding by other multi-step intercalating ligands and proteins.     

Mixed Mode of Ligand-DNA Binding Results in S-Shaped Binding Curves

Abstract

S-shaped binding curves often characterize interactions of ligands with nucleic acid molecules as analyzed by different physicochemical and biophysical techniques. S-shaped experimental binding curves are usually interpreted as indicative of the positive cooperative interactions between the bound ligand molecules. This paper demonstrates that S-shaped binding curves may occur as a result of the “mixed mode” of DNA binding by the same ligand molecule. Mixed mode of the ligand-DNA binding can occur, for example, due to 1) isomerization or dimerization of the ligands in solution or on the DNA lattice, 2) their ability to intercalate the DNA and to bind it within the minor groove in different orientations. DNA- ligand complexes are characterized by the length of the ligand binding site on the DNA lattice (so-called “multiple-contact” model). We show here that if two or more complexes with different lengths of the ligand binding sites could be produced by the same ligand, the dependence of the concentration of the complex with the shorter length of binding site on the total concentration of ligand should be S-shaped. Our theoretical model is confirmed by comparison of the calculated and experimental CD binding curves for bis-netropsin binding to poly(dA-dT) poly(dA-dT). Bis-netropsin forms two types of DNA complexes due to its ability to interact with the DNA as monomers and trimers. Experimental S-shaped bis-netropsin-DNA binding curve is shown to be in good correlation with those calculated on the basis of our theoretical model. The present work provides new insight into the analysis of ligand-DNA binding curves.     

Volume and hydration changes of DNA-ligand interactions

We report the volumetric and other thermodynamic properties of ethidium bromide (EB), propidium iodide (PI) and daunomycin (DAU) intercalating with poly(dA).poly(dT), poly[d(A-T)].poly[d(A-T)], and poly[d(G-C)].poly[d(G-C)], respectively, as well as minor groove binder Hoechst 33258 binding with poly[d(A-T)].poly[d(A-T)]. The data were obtained using fluorescence titration and hydrostatic pressure measurements. Our thermodynamic data are combined with enthalpies from literature reports to analyze the thermodynamic characteristics of the different interactions. The differences are interpreted based on three processes related to hydration: I. burial of non-polar hydrophobic solvent accessible surface, II. burial of polar surface and formation of solute-solute H-bonds, and III. disruption of "structural" hydration. Sequence dependent conformational changes may also be important when comparing ligand binding to different DNA sequences. We conclude that a combination of different thermodynamic parameters, especially volume change, is essential in order to understand the role of hydration in the energetics of DNA-ligand interactions.     

Complexation of anthracycline drugs with DNA in the presence of caffeine

The competitive binding of anthracycline antitumour drugs, [daunomycin (DAU), doxorubicin (DOX) or nogalamycin (NOG)], with caffeine (CAF) to a model DNA oligomer has been investigated by 500 MHz ¹H NMR spectroscopy under physiological solution conditions. The method depends on the stepwise analysis of one-component (self-association), two-component (hetero-association and DNA complexation) and three-component interactions, in order to de-convolute the overall binding of the anthracycline antibiotic and CAF to DNA into two competing processes, viz. hetero-association of the antibiotic-CAF (‘interceptor’ action of CAF) and CAF–DNA complexation (‘protector’ action of CAF). It is found that the complexation of DAU with DNA in the presence of CAF is mainly affected by the CAF–DNA complexation, whereas the binding of either DOX or NOG to DNA is affected approximately equally by both the CAF–DNA complexation and CAF-antibiotic hetero-association. Quantitative evaluation of the three-component mixture of drug–CAF–DNA has enabled the proportion of the antibiotic displaced from DNA on addition of CAF to be calculated over a large range of CAF concentration, which may provide a quantitative basis for the change in anthracycline-related toxicity on addition of CAF.     

Cooperative effects during the binding of large ligands with DNA. Non-contact interaction between adsorbed ligands

Equations are derived to describe the cooperative binding of large ligands to DNA. A mathematical approach is developed which enables one to give a simple probabilistic interpretation of binding equations and to solve them in the general case when long-range interactions are allowed between bound ligands. These interactions can be mediated by conformation changes induced in the DNA in the course of binding process and transformed over some distances beyond the DNA region immediately covered by a bound ligand molecule (allosteric effect of DNA). Interactions between ligand molecules can be formally described in terms of model potential characterizing pairwise interactions between bound ligands. A procedure is developed which allows one to determined the form of such potential from experimentally measured binding isotherms. It is based on a comparison of experimental binding isotherms with the appropriate curves calculated for the case of non-interacting ligands.     

NMR determination of the hetero-association of phenanthridines with daunomycin and their competitive binding to a DNA oligomer

500 MHz (1)H NMR spectroscopy has been used to determine thermodynamic and structural information on the hetero-association of daunomycin (DAU) with the phenanthridine mutagenic dyes ethidium bromide (EB) and propidium iodide (PI). The NMR complexation data have been analysed by a statistical-thermodynamic model which takes into account indefinite association for both the self-association of the drugs and their hetero-association. The results have been used to estimate the effect of the side chains of the phenanthridines on the competitive binding between DAU and the mutagens with DNA. Knowledge of the equilibrium constants for self-association of the phenanthridines and DAU, their hetero-association and their complexation with a DNA fragment, the deoxytetranucleotide 5'-d(TpGpCpA), enabled the relative content of each of the EB-DAU, PI-DAU, EB-DAU-d(TGCA) and PI-DAU-d(TGCA) complexes to be calculated as a function of drug concentration in mixed solutions. The results provide some insight into the molecular basis of the action of combinations of biologically-active molecules. When intercalating drugs are used in combination, it is found that the decrease in binding of drug or mutagen with DNA is due both to formation of drug-mutagen hetero-association complexes in the mixed solution and to competition for the binding sites by the aromatic molecules; the relative importance of each process depends on the molecular properties of the drug or mutagen molecules being considered. Thus, the longer branched side chain of PI and the electrostatic contribution of the extra positive charge of the molecule compared with the ethyl group of EB results in lower affinity for self-association of PI molecules and their hetero-association with DAU, but increases the degree of binding of PI with DNA.     

Quantitative and sequence-specific analysis of DNA-ligand interaction by means of fluorescent intercalator probes

A novel method of analysis of double-stranded DNA-ligand interaction is presented. The interaction is monitored by the fluorescence of a DNA bis-intercalator oxazole homodimer YoYo-3. The fluorescence intensity or its decay time reflects the modification of the DNA double helix. The DNA sequence is scanned by hybridization with short oligomers having consecutively overlapping complementary sequences to analyse the sequence specificity of binding. In our experiments we used as ligands the minor groove binders netropsin, SN6999 (both with AT-preference), the GC-specific ligand chromomycin A3 as well as the derivative SN6113 (non-specific interaction), which displace the bis-intercalator YoYo-3 or influence the duplex structure in such away that the fluorescence intensity and lifetime decrease in comparison to a ligand-free screening. The changes of fluorescence emission clearly define the binding motif and indicate minor groove interactions with a reduced DNA binding site. Titration of the ligand quantitatively characterizes its binding by determining the dependence of the binding constant on the oligonucleotide sequence.     

Molecular events associated with CD4-mediated Down-regulation of LFA-1-dependent adhesion.

We have previously shown that CD4 ligand binding inhibits LFA-1-dependent adhesion between CD4+ T cells and B cells in a p56(lck)- and phosphatidylinositol 3-kinase (PI3-kinase)-dependent manner. In this work, downstream events associated with adhesion inhibition have been investigated. By using HUT78 T cell lines, CD4 ligands were shown to induce a dissociation of LFA-1 from cytohesin, a cytoplasmic protein known to bind LFA-1 and to enhance the affinity/avidity of LFA-1 for its ligand ICAM-1. A dissociation of PI3-kinase from cytohesin is also observed. In parallel, we have found that CD4 ligand binding induced a redistribution of PI3-kinase and of the tyrosine phosphatase SHP-2 to the membrane and induced a transient formation of protein interactions including PI3-kinase; an adaptor protein, Gab2; SHP-2; and a SH2 domain-containing inositol phosphatase, SHIP. By using antisense oligonucleotides or transfection of transdominant mutants, down-regulation of adhesion was shown to require the Gab2/PI3-kinase association and the expression of SHIP and SHP-2. We therefore propose that CD4 ligands, by inducing these molecular associations, lead to sustained local high levels of D-3 phospholipids and possibly regulate the cytohesin/LFA-1 association.     

Interaction of effecting ligands with lac repressor and repressor-operator complex.

The equilibrium association constants for the binding of a wide variety of effecting ligands of the lac repressor were measured by equilibrium dialysis. Also, detailed investigations of the apparent rate of dissociation of repressor-operator comples as a function of ligand concentration were carried out for several inducers and anti-inducers. The affinity of repressor-ligand comples for operator DNA was evaluated from the specific rate constants at saturating concentrations of effecting ligand. By fitting the experimental data depicting the functional dependence of the rate of dissociation upon ligand concentrations to calculated curves, assuming simple models of the induction mechanism, the equilibrium association constant for the binding of effecting ligand to repressor-operator comples was determined. Inducers reduce the affinity of lac repressor for operator DNA by a factor of approximately 1000 under standard conditions; the extent of destabilization depends on Mg2+ ion concentration. Anti-inducers increase the affinity of repressor for operator at most a factor of five. Only one neutral ligand, which binds to repressor without altering the stability of repressor-operator comples, was found. No homotropic or heterotropic interactions in the binding of effecting ligands either to repressor or to repressor-operator complex are evident.     

Caffeine, pentoxifylline and theophylline form stacking complexes with IQ-type heterocyclic aromatic amines

Methylxanthines (MTX), in particular caffeine (CAF), are known as the most widely consumed alkaloids worldwide. Many accumulated statistical data indicate the protective effect of CAF intake against several types of cancer. One of the possible explanations of this phenomenon is direct non-covalent interaction between CAF and aromatic mutagen/carcinogen molecules through stacking (π-π) complexes formation. Here we demonstrate that CAF and other MTX, pentoxifylline (PTX) and theophylline (TH), form stacking complexes with carcinogenic imidazoquinoline-type (IQ-type) food-borne heterocyclic aromatic amines (HCAs). We estimated neighborhood association constants (K_AC of the order of magnitude of 10² M⁻¹) in neutral and acidic environment and enthalpy changes (ΔH values between −15.1 and −39.8 kJ/mol) for these interactions using UV-Vis spectroscopy, calculations based on thermodynamical model of mixed aggregation and titration microcalorimetry. Moreover, using Ames test with Salmonella typhimurium TA98 strain and recently developed mutagenicity assay based on bioluminescence of Vibrio harveyi A16 strain, we demonstrated a statistically significant reduction in HCAs mutagenic activity in the presence of MTX.     

Equilibrium binding of derivatives of the carcinogen, benzo(a)pyrene, to DNA. Thermodynamic analysis

The physical binding of polycyclic aromatic hydrocarbon derivatives which are ultimate carcinogens to DNA may play a role in the formation of covalent DNA adducts by these compounds or in the detoxification of the compounds via DNA-catalyzed hydrolysis. Previous studies of DNA-binding interactions of derivatives of benzo(a)pyrene (BP) have been confined to low r values (r - ligands bound/base pair). We have now applied the Scatchard formalism (as modified to include neighbor exclusion) to the spectrophotometric determination of the binding of two derivatives of BP, trans - 9,10 - dihydroxydihydro - BP and 7r,8t - dihydroxy-9t,10t-oxy-7,8, 9,10-tetrahydro-BP, to double-stranded DNA at reasonably high r values. Exclusion parameters, binding constants, and thermodynamic parameters are all within the ranges found for other intercalants. Although these ligands are uncharged, the binding exhibits significant ionic strength dependence which can be rationalized (partially) by polyelectrolyte theory. Using the measured ionic strength dependence, a thermodynamic association constant, independent of ionic interactions, can be calculated which is very close to the calculated thermodynamic association constants for ethidium and proflavine.     

Theoretical calculations of the helix–coil transition of DNA in the presence of large,cooperatively binding ligands

Theoretical calculations are conducted on the helix–coil transition of DNA, in the presence of large, cooperatively binding ligands modeled after the DNA-binding proteins of current biological interest. The ligands are allowed to bind both to helx and to coil, to cover up any number of bases or base pairs in the complex, and to interact cooperatively with their nearest neighbors. The DNA is treated in the infinite homogeneous Ising model approximation, and all calculations are done by Lifson's method of sequence-generating functions. DNA melting curves are calculated by computer in order to expolore the effects on the transition of ligand size, binding constant, free activity, and ligand–ligand cooperativity. The calculations indicate that (1) at the same intrinsic free energy change per base pair of the complexes, small ligands, for purely entropic reasons, are more effective than are large ligands in shifting the DNA melting temperature; (2) the response of the DNA melting temperature to increased ligand binding constant K and/or free ligand activity L is adequately represented at high values of KL (but not at low KL) by a simple independent site model; (3) if curves are calculated with the total amount of added ligand remaining constant and the free ligand activity allowed to vary throughout the transition, biphasic melting curves can be obtained in the complete absence of ligand–ligand cooperativity. In an Appendix, the denaturation of poly[d(A-T)] in the presence of the drug, netropsin, is used to verify some features of the theory and to illustrate how the theory can be used to obtain numerical estimates of the ligand binding parameters from the experimental melting curves.