Barira Islam

Conformational dynamics of the human propeller telomeric DNA quadruplex on a microsecond time scale

The human telomeric DNA sequence with four repeats can fold into a parallel-stranded propeller-type topology. NMR structures solved under molecular crowding experiments correlate with the crystal structures found with crystal-packing interactions that are effectively equivalent to molecular crowding. This topology has been used for rationalization of ligand design and occurs experimentally in a number of complexes with a diversity of ligands, at least in the crystalline state. Although G-quartet stems have been well characterized, the interactions of the TTA loop with the G-quartets are much less defined. To better understand the conformational variability and structural dynamics of the propeller-type topology, we performed molecular dynamics simulations in explicit solvent up to 1.5 μs. The analysis provides a detailed atomistic account of the dynamic nature of the TTA loops highlighting their interactions with the G-quartets including formation of an A:A base pair, triad, pentad and hexad. The results present a threshold in quadruplex simulations, with regards to understanding the flexible nature of the sugar-phosphate backbone in formation of unusual architecture within the topology. Furthermore, this study stresses the importance of simulation time in sampling conformational space for this topology.

A G-quadruplex-binding compound showing anti-tumour activity in an in vivo model for pancreatic cancer.

We report here that a tetra-substituted naphthalene-diimide derivative (MM41) has significant in vivo anti-tumour activity against the MIA PaCa-2 pancreatic cancer xenograft model. IV administration with a twice-weekly 15 mg/kg dose produces ca 80% tumour growth decrease in a group of tumour-bearing animals. Two animals survived tumour-free after 279 days. High levels of MM41 are rapidly transported into cell nuclei and were found to accumulate in the tumour. MM41 is a quadruplex-interactive compound which binds strongly to the quadruplexes encoded in the promoter sequences of the BCL-2 and k-RAS genes, both of which are dis-regulated in many human pancreatic cancers. Levels of BCL-2 were reduced by ca 40% in tumours from MM41-treated animals relative to controls, consistent with BCL-2 being a target for MM41. Molecular modelling suggests that MM41 binds to a BCL-2 quadruplex in a manner resembling that previously observed in co-crystal structures with human telomeric quadruplexes. This supports the concept that MM41 (and by implication other quadruplex-targeting small molecules) can bind to quadruplex-forming promoter regions in a number of genes and down-regulate their transcription. We suggest that quadruplexes within those master genes that are up-regulated drivers for particular cancers, may be selective targets for compounds such as MM41.

Can We Execute Reliable MM-PBSA Free Energy Computations of Relative Stabilities of Different Guanine Quadruplex Folds?

The self-assembly and stability of DNA G-quadruplexes (GQs) are affected by the intrinsic stability of different GpG base steps embedded in their G-quartet stems. We have carried out MD simulations followed by MM-PBSA (molecular mechanics Poisson-Boltzmann surface area) free energy calculations on all the experimentally observed three-quartet intramolecular human telomeric GQ topologies. We also studied antiparallel GQ models with alternative syn-anti patterns of the G-quartets. We tested different ions, dihedral variants of the DNA force field, water models, and simulation lengths. In total, ∼35 μs of simulations have been carried out. The systems studied here are considerably more complete than the previously analyzed two-quartet stems. Among other effects, our computations included the stem-loop coupling and ion-ion interactions inside the stem. The calculations showed a broad agreement with the earlier predictions. However, the increase in the completeness of the system was associated with increased noise of the free energy data which could be related, for example, to the presence of long-lived loop substates and rather complex dynamics for the two bound ions inside the G-stem. As a result, the MM-PBSA data were noisy and we could not improve their quantitative convergence even by expanding the simulations to 2.5 μs long trajectories. We also suggest that the quality of MM-based free energy computations based on MD simulations of complete GQs is more sensitive to the neglect of explicit polarization effects, which, in real systems, are associated with the presence of multiple closely spaced ions inside the GQs. Thus, although the MM-PBSA procedure provides very useful insights that complement the structural-dynamics data from MD trajectories of GQs, the method is far from reaching quantitative accuracy. Our conclusions are in agreement with critical assessments of the MM-PBSA methodology available in contemporary literature for other types of problems.

Selectivity of major isoquinoline alkaloids from Chelidonium majus towards telomeric G-quadruplex: A study using a transition-FRET (t-FRET) assay

BACKGROUND Natural bioproducts are invaluable resources in drug discovery. Isoquinoline alkaloids of Chelidonium majus constitute a structurally diverse family of natural products that are of great interest, one of them being their selectivity for human telomeric G-quadruplex structure and telomerase inhibition. METHODS The study focuses on the mechanism of telomerase inhibition by stabilization of telomeric G-quadruplex structures by berberine, chelerythrine, chelidonine, sanguinarine and papaverine. Telomerase activity and mRNA levels of hTERT were estimated using quantitative telomere repeat amplification protocol (q-TRAP) and qPCR, in MCF-7 cells treated with different groups of alkaloids. The selectivity of the main isoquinoline alkaloids of Chelidonium majus towards telomeric G-quadruplex forming sequences were explored using a sensitive modified thermal FRET-melting measurement in the presence of the complementary oligonucleotide CT22. We assessed and monitored G-quadruplex topologies using circular dichroism (CD) methods, and compared spectra to previously well-characterized motifs, either alone or in the presence of the alkaloids. Molecular modeling was performed to rationalize ligand binding to the G-quadruplex structure. RESULTS The results highlight strong inhibitory effects of chelerythrine, sanguinarine and berberine on telomerase activity, most likely through substrate sequestration. These isoquinoline alkaloids interacted strongly with telomeric sequence G-quadruplex. In comparison, chelidonine and papaverine had no significant interaction with the telomeric quadruplex, while they strongly inhibited telomerase at transcription level of hTERT. Altogether, all of the studied alkaloids showed various levels and mechanisms of telomerase inhibition. CONCLUSIONS We report on a comparative study of anti-telomerase activity of the isoquinoline alkaloids of Chelidonium majus. Chelerythrine was most effective in inhibiting telomerase activity by substrate sequesteration through G-quadruplex stabilization. GENERAL SIGNIFICANCE Understanding structural and molecular mechanisms of anti-cancer agents can help in developing new and more potent drugs with fewer side effects. Isoquinolines are the most biologically active agents from Chelidonium majus, which have shown to be telomeric G-quadruplex stabilizers and potent telomerase inhibitors.

Exploring the Dynamics of Propeller Loops in Human Telomeric DNA Quadruplexes Using Atomistic Simulations

We have carried out a series of extended unbiased molecular dynamics (MD) simulations (up to 10 μs long, ∼162 μs in total) complemented by replica-exchange with the collective variable tempering (RECT) approach for several human telomeric DNA G-quadruplex (GQ) topologies with TTA propeller loops. We used different AMBER DNA force-field variants and also processed simulations by Markov State Model (MSM) analysis. The slow conformational transitions in the propeller loops took place on a scale of a few μs, emphasizing the need for long simulations in studies of GQ dynamics. The propeller loops sampled similar ensembles for all GQ topologies and for all force-field dihedral-potential variants. The outcomes of standard and RECT simulations were consistent and captured similar spectrum of loop conformations. However, the most common crystallographic loop conformation was very unstable with all force-field versions. Although the loss of canonical γ-trans state of the first propeller loop nucleotide could be related to the indispensable bsc0 α/γ dihedral potential, even supporting this particular dihedral by a bias was insufficient to populate the experimentally dominant loop conformation. In conclusion, while our simulations were capable of providing a reasonable albeit not converged sampling of the TTA propeller loop conformational space, the force-field description still remained far from satisfactory.

Effect of Monovalent Ion Parameters on Molecular Dynamics Simulations of G-Quadruplexes

G-quadruplexes (GQs) are key noncanonical DNA and RNA architectures stabilized by desolvated monovalent cations present in their central channels. We analyze extended atomistic molecular dynamics simulations (∼580 μs in total) of GQs with 11 monovalent cation parametrizations, assessing GQ overall structural stability, dynamics of internal cations, and distortions of the G-tetrad geometries. Majority of simulations were executed with the SPC/E water model; however, test simulations with TIP3P and OPC water models are also reported. The identity and parametrization of ions strongly affect behavior of a tetramolecular d[GGG]4 GQ, which is unstable with several ion parametrizations. The remaining studied RNA and DNA GQs are structurally stable, though the G-tetrad geometries are always deformed by bifurcated H-bonding in a parametrization-specific manner. Thus, basic 10-μs-scale simulations of fully folded GQs can be safely done with a number of cation parametrizations. However, there are parametrization-specific differences and basic force-field errors affecting the quantitative description of ion-tetrad interactions, which may significantly affect studies of the ion-binding processes and description of the GQ folding landscape. Our d[GGG]4 simulations indirectly suggest that such studies will also be sensitive to the water models. During exchanges with bulk water, the Na+ ions move inside the GQs in a concerted manner, while larger relocations of the K+ ions are typically separated. We suggest that the Joung-Cheatham SPC/E K+ parameters represent a safe choice in simulation studies of GQs, though variation of ion parameters can be used for specific simulation goals.

Structural Dynamics of Lateral and Diagonal Loops of Human Telomeric G-Quadruplexes in Extended MD Simulations

The NMR solution structures of human telomeric (Htel) G-quadruplexes (GQs) are characterized by the presence of two lateral loops complemented by either diagonal or propeller loops. Bases of a given loop can establish interactions within the loop as well as with other loops and the flanking bases. This can lead to a formation of base alignments above and below the GQ stems. These base alignments are known to affect the loop structures and relative stabilities of different Htel GQ folds. We have carried out a total of 217 μs of classical (unbiased) molecular dynamics (MD) simulations starting from the available solution structures of Htel GQs to characterize structural dynamics of the lateral and diagonal loops, using several recent AMBER DNA force-field variants. As the loops are involved in diverse stacking and H-bonding interactions, their dynamics is slow, and extended sampling is required to capture different conformations. Nevertheless, although the simulations are far from being quantitatively converged, the data suggest that multiple 10 μs-scale simulations can provide a quite good assessment of the loop conformational space as described by the force field. The simulations indicate that the lateral loops may sample multiple coexisting conformations, which should be considered when comparing simulations with the NMR models as the latter include ensemble averaging. The adenine-thymine Watson-Crick arrangement was the most stable base pairing in the simulations. Adenine-adenine and thymine-thymine base pairs were also sampled but were less stable. The data suggest that the description of lateral and diagonal GQ loops in contemporary MD simulations is considerably more realistic than the description of propeller loops, though definitely not flawless.