STRUCTURAL STUDIES OF FLAVONOIDS AND THEIR INTERACTION WITH DUPLEX AND QUADRUPLEX DNA

Abstract

Over the last two decades many strategies have been planned to design specific drugs for rare diseases like cancer to target their action at the DNA level. Advancements in our understanding of the interaction of small molecules with DNA have opened the doors for "rational" drug design. Special methods have now been developed to give accurate account of the precise location of ligand-DNA adducts on target DNA. We are now in a position to think of designing or explore ligands of our choice that recognize particular sequences of base pairs. Plants have evolved a diverse set of molecules/ligands that bind to DNA in a variety of ways, but with the common ability to act as potent inhibitors of DNA transcription and replication. As a consequence, plant products like flavonoids have been of considerable interest as potential anti cancer agents. Many compounds have been added to this list in the search for more potent drugs for use in chemotherapy in view of pronounced cytotoxicity of other synthetic drugs, resistance towards tumor cell lines and difference in their neoplastic potency. This work will allow us to enter into a new era of cancer therapy using ligand of the natural origin, available in our daily diets and flavonoids are one of them. A substantial body of research has been directed towards understanding the molecular basis of action and DNA sequence specificity for binding, by identifying the preferred binding sequences of many key drugs with DNA. Structural tools such as X-ray crystallography and NMR spectroscopy, coupled with molecular modeling techniques have considerable impact in advancing our understanding of the microscopic structural homogeneity of DNA and the molecular basis for drug-DNA interactions. In present work, UV-Visible and fluorescence spectroscopy, Nuclear Magnetic Resonance Spectroscopy (NMR) followed by restrained Molecular Dynamics (rMD) simulations and theoretical studies using Density Functional method (DFT) by Gaussian 98 and 03 have been used. These studies are supported by Diffusion Ordered Spectroscopy (DOSY), Electron Spray Ionization Mass Spectrometry (ESI-MS) and Fluorescence Life-Time Measurements to investigate about the binding mode of flavonoids with different duplex and quadruplex DNA sequences. The Ph.D. thesis work has been reported in the form of seven chapters. Chapter 1 contains introduction of the subject, a comprehensive review of the literature and scope of thesis. Chapter 2 deals with the materials and methods used. The methods of detailed competition dialysis, UV-Visible, fluorescence spectroscopy and Electron Spray Ionization Mass Spectrometry (ESI-MS) to reveal DNA binding for their sequence specificity and selectivity, Nuclear Magnetic Resonance Spectroscopy- ID NMR, DQF COSY, TOCSY, 'H - 'H NOESY for the proton assignment; HSQC ('H-13C) and HMBC ('H-13C) for the carbon assignment; *H -31P HMBC for the phosphorus assignment, 31P - 31P NOESY; UV-Visible and Fluorescence spectroscopy studies are done as a function of drug to DNA ratio; Diffusion Ordered Spectroscopy (DOSY) and Electron Spray Ionization Mass Spectrometry (ESI-MS) are done to understand the self-association; Fluorescence Life-Time Measurements are explained. The strategies used for restrained energy minimization, restrained Molecular Dynamics simulations and quantum mechanical calculations involving GIAO method (for chemical shift calculation) and DFT method (for optimization) are also discussed. Chapter 3deals with the calculation of chemical shift of *H and 13C resonances in Nuclear Magnetic Resonances (NMR) spectra of representative molecules of flavone (Crysin, Apigenin, Luteolin and Acacetin), flavanol (Galangin, Kaemferol, Quercetin and Myricetin) and flavanone (Hesperetin and Naringenin) members of flavonoid group using the Gauge-Invariant Atomic Orbital (GIAO) method as implemented in Gaussian 98 for gas phase and Gaussian 03 for solvent phase, using Density Functional Theory (DFT) employing B3LYP exchange correlation at 6-311G** level of basis set. Molecular properties like electronic charges, dipole and energies and structural parameters like bond length, bond angles and torsional angles were calculated for all the hydroxyl substituted flavonoids in gas phase and in solvent DMSO, using the same basis set. These properties were correlated with their functional aspects. Based on the results, it was ascertained that acacetin (5,7-Dihydroxy-4'-methoxyflavone) and galangin (3,5,7-trihydroxy flavone) is non intercalator with DNA and other flavones like apigenin, luteolin and flavanols like kaemferol, quercetin and myricetin are found to be capable of both intercalation and topoisomerase poisoning. None of the flavanone derivatives, which have a saturated 2,3- bond, are active. It may be proposed that a minimal flavone structure for DNA intercalation and topoisomerase poisoning is required to be 3', 4'-dihydroxyflavone. Chapter 4 deals with the NMR chemical shift, rMD simulations and quantum mechanical calculations of luteolin, quercetin, rutin and genistein (Fig. 1). In this chapter, the structural and electronic properties of these flavonoids have been studied using Density Functional Theory (DFT) employing B3LYP exchange correlation. The chemical shift of *H and 13C resonances in Nuclear Magnetic Resonance spectra have been calculated using Gauge-Invariant Atomic Orbital (GIAO) method as implemented in Gaussian 98 for gas phase and Gaussian 03 for solvent DMSO and compared with experimental data. The optimized solution structure of drugs was obtained by rMD simulation using inter-proton distance constraints obtained from 2D ROESY spectra. It was observed that A and C ring are planar in these flavonoids and the B ring of the quercetin and rutin rotate around C2'-C1 '-C2 -01 torsional angle to go in to two different conformation, anti and syn. The presence of anti and syn conformers have also been confirmed by solid state NMR and DFT studies. Similar effect has been also observed in case of genistein but around C6'-C1'-C3 -C4 torsional angle. A variation in flexibility of adopting two different conformations in these flavonoids may be related to a differential efficacy in biological activity and hence antioxidant or anticancer action. Chapter 5 deals with the studies on self-association of luteolin, quercetin, rutin and genistein by rMD approach using one and two dimensional Nuclear Magnetic Resonance Spectroscopy supported by Absorption, Fluorescence, Diffusion Ordered Spectroscopy and Mass Spectrometry. Self association is a process which competes with binding to DNA and in formation of hetero-complexes. Concentration (1-10 uM) as well temperature (278-353 K) dependent one-dimensional proton NMR experiments are done and the change in chemical shift shows the presence of self association in all the four molecules. The two-dimensional NOESY studies show several intra-molecular and intermolecular inter-proton connectivities suggesting specific dimer patterns involving hydrogen bonding. Absorption, emission and diffusion ordered spectroscopy demonstrate formation of self aggregates and accordingly association constant (Ka) is calculated. From DOSY experiment, it has been found that the apparent diffusion rate is fastest for luteolin and least for rutin. ESI-MS studies also prove the presence of dimer and higher aggregates which is also confirmed by isolation and fragmentation of dimer species by MS-MS analysis. The structural differences between these molecules have been correlated to biological action. Chapter 6 reports our attempt to investigate the interaction of four flavonoids, viz. luteolin, quercetin, genistein and rutin, having a similar structure but differing only in number and /or distribution of OH groups, with different duplex and telomeric G-quadruplex DNA sequences. Duplex DNA viz. calf thymus DNA, poly d(A-T), poly d(G-C), poly d(A .T) and quadruplex DNA d-(TTAGGG)4, d-(TTAGGGT)4, d-(TTGGGGT)4 are used for titration with all these four flavonoids using competition dialysis assay, UV- visible and fluorescence spectroscopy. The life time fluorescence and mass spectrometric experiments also substantiate the effective binding of these four flavonoids with the G-quadruplex DNA. Competitive binding assay results show that relative affinity with different duplex and quadruplex DNA. The flavonoids shows maximum binding with calf thymus DNAand the trend for binding is quercetin > rutin > genistein > luteolin. Further, flavonoids prefer to bind with poly d-(A-T) rather to bind with poly d-(G-C). Similarly, with quadruplex DNA d-(TTAGGG)4 , these flavonoids show binding in the order of quercetin > rutin > luteolin > genistein while for other two sequences not much difference was found. Further, in absorption spectroscopic studies, the % hyperchromicity of these flavonoids with calf thymus DNAdecreases in the following order: Luteolin > Genistein > Rutin > Quercetin; with poly d (A-T): Genistein > Rutin > Luteolin > Quercetin; while with poly d(A) it decreases in the order: Genistein > Rutin > Luteolin > Quercetin; with poly d(G-C) it decreases in the order: Genistein >Rutin > Luteolin > Quercetin; with (TTAGGGT)4 it decreases in the order: Luteolin > Rutin > Genistein > Quercetin ands with (TTGGGGT)4 decreases in the order: Rutin > Luteolin > Genistein > Quercetin. In the negative-ion ESI-MS spectra, peaks corresponding to the quadruplex DNA alone as well as withpotassium adducts andwithdifferent flavonoids having stoichiometries (1:1) and (1:2) are clearly observed. Chapter 7 deals with a detailed phosphorus and proton NMR study of binding of quercetin with human telomeric G-quadruplex sequence d-(TTAGGGT)4. Titration studies are performed upto drug to DNA duplex ratio of2.0 at 283, 298 and 318 K. In 31P NMR, the 6 phosphate group resonances are observed in d-(TTAGGGT)4, and further addition of quercetin up to 1:1 complex induces significant upfield shift with appearance of new set of resonance showing change in phosphodiester backbone. In proton experiments, 2D NOESY 1:1, 1.5:1 and 2.0:1 flavonoids-G-quadruplex complex yields several intra-molecular and inter-molecular contacts. We do not see evidence of thymine imino protons even at 283 K, indicating that quercetin binding does not stabilize or induce T-tetrad formation. Further, absence of most of the sequential connectivities and the intra-base pair contacts show that DNA quadruplex is not intact. The drug is intercalated into the base pair of DNA, placed close to Tl and T2 base pairs. Further, large change in proton chemical shift of T7H6 and presence of sequential connectivity between G6H1'- T7H6 shows the binding of another molecule of quercetin below T7 base. A model was built using the inter-proton distances obtained from NOESY cross peaks by rMD simulations on complex. The torsional angles and inter-proton distances obtained from proton NMR experiments, exchange of bound and free drug by 31P NMR experiments, along with the rMD simulations of the structure of drug-DNA complex show that these flavonoids intercalate between TlpT2 base pairs and end stack G6pT7 of DNA and stabilize the G-quadruplex complex. Diffusion Ordered Spectroscopy (DOSY) studies provide evidence of done to see the formation of intercalated complex and the spectra show two different populations of bound and free DNA as compared to that alone DNA. Chapter 8 summarizes the results obtained and their implications in understanding the arrangement/ substitution of hydroxyl group on flavonoids and their molecular basis of action for targeting DNA for anti cancer therapy.