Conformational studies and solvent-accessible surface area analysis of known selective DNA G-Quadruplex binders

Biochimie 93 (2011) 1267e1274

Contents lists available at ScienceDirect

Biochimie journal homepage: www.elsevier.com/locate/biochi

Research paper

Conformational studies and solvent-accessible surface area analysis of known selective DNA G-Quadruplex binders Stefano Alcaro*, Anna Artese, Giosuè Costa, Simona Distinto, Francesco Ortuso, Lucia Parrotta Dipartimento di Scienze Farmacobiologiche, Laboratorio di Chimica Farmaceutica, Università degli Studi "Magna Græcia", Complesso “Ninì Barbieri”, Roccelletta di Borgia, 88021 (Catanzaro), Italy

a r t i c l e i n f o

a b s t r a c t

Article history: Received 1 March 2011 Accepted 14 June 2011 Available online 16 June 2011

Human telomeres are comprised of d(TTAGGG) repeats involved in the formation of G-quadruplex DNA structures. Ligands that stabilize these G-quadruplex DNA structures are potential inhibitors of the cancer cell-associated enzyme telomerase. In human cells, telomerase adds multiple copies of the 50 GGTTAG-30 motif to the end of the G-strand of the telomere and in the majority of tumor cells it results over-expressed. Several structural studies have revealed a diversity of topologies for telomeric quadruplexes, as confirmed by the different conformations deposited in the Protein Data Bank. In recent years an increasing number of chemically diverse telomerase inhibitors have been identified, including both natural and synthetic compounds. Thus telomerase has been regarded as one of the most attractive targets in cancer treatment. In this manuscript, with the aim to rationalize the different experimental activities of known telomerase inhibitors, a computational study was carried out to investigate their conformational properties and the relationships between the target affinity and the ligands solventaccessible surface area. Among the analyzed different scaffolds of G-quadruplex binders, such a descriptor provided helpful preliminary information to discriminate end-stacking ligand binding affinities, revealing itself as a useful predictive tool in drug design and lead optimization processes. Ó 2011 Published by Elsevier Masson SAS.

Keywords: G-quadruplex Conformational search Solvent-accessible surface area Boltzmann population TRAP

1. Introduction The telomeres of human cells protect chromosomal ends from fusion events; they have a length from 3 kb to 15 kb and are composed of tandem repeats of the sequence 50 -GGTTAG-30 with 30 overhang of the G-strand essential for structural and functional roles [1]. In human somatic cells, telomere length decreases with each cell division event [2]. The infinite extension of telomeric ends is associated with an aberrant cellular proliferation [3]. Cancer is characterized by such an immortalized state, in most cases by the activation of the reverse transcriptase enzyme telomerase, which is expressed in 85e90% of cancer cells and not significantly expressed in somatic tissue [4]. Since telomere maintenance is associated with the unlimited proliferative potential of cancer cells, most telomere related antitumor strategies target the telomerasedependent mechanism. Several studies have demonstrated telomerase as a reliable marker [5] for some cancers as well as a target of inhibitors of immortal cell growth [6]. In presence of certain cations, the eukaryotic chromosomes at the ends of telomeric DNA G-rich sequences can associate together * Corresponding author. Tel.: þ39 0961 3694197. E-mail address: [email protected] (S. Alcaro). 0300-9084/$ e see front matter Ó 2011 Published by Elsevier Masson SAS. doi:10.1016/j.biochi.2011.06.014

to form particular four-stranded conformations known as quadruplexes, tetraplexes or G4 structures [7]. Quadruplexes can be formed from one, two or four separate strands of DNA (or RNA) and display a large variety of topologies, widely analyzed by means of structural studies [3]. All quadruplexes are characterized by a repeating motif known as G-quartet, also defined G-tetrad, characterized by a high tendency to produce self-stacking [8]. Quadruplexes have a requirement for metal ions, especially the alkali metals [9], placed in the interior channel formed at the center of each quadruplex. The loops observed in the different quadruplex topologies can be diagonal, lateral or chain-reversal. In particular the number of G-quartets, the loop length, sequence and sometimes the nature of the cation are related to the plausible diverse structural arrangements. By definition chain-reversal loops connect two strands in the same parallel orientation, whereas diagonal and lateral loops connect chains in opposing, antiparallel orientations [10]. In presence of Naþ and Kþ ions, six different main conformations of the monomolecular human telomeric sequence have been identified and deposited in the Protein Data Bank (PDB) [11] respectively with the codes 143D (antiparallel, Naþ) [12], 1KF1 (parallel, Kþ) [13], 2HY9 (hybrid-1, Kþ) [14], 2JPZ (hybrid-2, Kþ) [15], 2JSL (hybrid-1, Kþ) [16], and 2JSM (hybrid-2, Kþ) [16].

1268

S. Alcaro et al. / Biochimie 93 (2011) 1267e1274

Such DNA conformations have provided novel targets for designing compounds with improved specificity. G-quadruplex selective binders or stabilizers may interfere with telomere conformation and elongation. Synthetic small molecules were the first ligands identified to fold G-quadruplex and indirectly inhibit telomerase activity [17]. Starting from this initial work, several quadruplex binding molecules have been identified as telomere blocking agents, but corresponded to weak inhibitors of telomerase extension [18,19a,b]. Structure-activity relationship studies and physico-chemical measurements highlighted large aromatic ring-systems bearing positively charged side chains as main structural features to be efficient G-quadruplex ligands. Secondary structure stabilization and thus telomerase inhibition implies pep stacking between planar aromatic molecules and the external guanine-rich face of Gquadruplex and electrostatic interactions of positively charged functions with the phosphate backbone of DNA [20e22]. Telomerase inhibitors of natural and synthetic origins are chemically diverse ligands such as telomestatin [23], ethidium derivatives [17], bisamido-anthraquinones [24], fluorenones [25a,b,c], acridones [26], acridines [27], perylene diimides [28], fluoroquinolones [29a,b], indoloquinolines [30], cryptolepines [31a,b,c], quindolines [32], porphyrins [33], phenanthrolines [34], triazines [35] and carbazole derivatives [36]. Among all these numerous heterocyclic systems, some substituted acridines excelled by their efficiency. Thus, the 3,6-bis(aminoalkyl) sidechain substituted acridine BRACO displayed potent G-quadruplex stabilization ability and Telomere Repeat Amplification Protocol (TRAP) inhibitory potential [37e39]. Replacement of the C-9 dimethylaminoaniline moiety by a difluorobenzyl group allowed pharmacological profile improvements [40]. Another class of compounds lately reported included perylene derivatives characterized by side chains ending with linear or cyclic amines. These molecules were capable of strong and selective recognition to the G-quadruplex showing a remarkable biological and pharmacological anticancer effects in living systems [41]. Even though experiments with the parent compound, ethidium bromide, had proven low selectivity toward G-quadruplexes, several ethidium derivatives, characterized by a high affinity and selectivity toward quadruplex DNA, were synthesized. Such an improvement was probably due to the aromatic surface of the derivatives, along with the positive charge inside the ring system, which could interact with negative charges in the G-quadruplex and also promote binding. Actually it has been shown that these ethidium derivatives not only stabilize G-quadruplex, but they promote its formation [42]. The binding of a ligand to its receptor is dependent on several chemical-physical factors. Many descriptors based approaches have been successfully applied to drug discovery process [43]. These methods correlate calculated properties derived from the chemical structure of examined compounds to experimentally determined biological activity and are generally indicated as Quantitative StructureeActivity Relationship (QSAR) methods. Molecular properties can be determined through experiments, but more often computational methods have been devised to calculate them from the topology of any given molecule. Although 3D-descriptors are conformationdependent and require more computational resources, 3D QSAR approaches are the most powerful and applied ones [44]. Among the alignment dependent methods, CoMFA is the most popular [45], however there are other 3D approaches which are independent from alignment. Some examples reported in literature consider distributions of molecular surfaces (MaP) [46], comparative molecular moment analysis (CoMMA) [47], and comparative spectra analysis (CoSA) [48] that apply molecular moments and molecular spectra as descriptors.

Scheme 1. The Boltzmann probability is computed as the ratio of the single microstate exponential and the partition equation. The value of pi is between 0 and 1.

In the present study we report the application of the well documented Averaged Solvent Accessible Surface Area (ASASA) descriptor as a tool to give direct insights into the role of the solvent in the binding event of known G-quadruplex binders exhibiting different conformational profiles. The concept of Solvent Accessible Surface Area (SASA) is of great importance in biological field. It was first introduced to evaluate the solvation effects in proteins and their complexes [49a,b,c,d]. It also became an important descriptor of the protein structure itself and has been used for the identification of binding sites and proteineprotein interfaces [50]. Moreover DNA ability to form double helix or unwind strongly depends on hydration [51]. In a recent work [52] the exploration of amino acid solvent accessibility was applied to DNA in order to give additional insights of DNA-waterprotein interactions. Furthermore the important role of the ASASA descriptor was also reported for the interaction of DNAsmall molecules [53,54], while only recently it was applied to investigate its significance for G-quadruplex binders [55e57]. 2. Materials and methods 2.1. Monte Carlo (MC) conformational search The conformational search was carried out by molecular mechanics techniques coupled to the analysis of the Solvent Accessible Surface Area (SASA) computed onto a defined planar aromatic system depending on the chemical nature of the analyzed ligands. In details, after building the 3-D structures of each compound, we proceeded with the exploration of their internal degrees of freedom by Monte Carlo (MC) randomization of any rotatable bond. 10,000 conformations were generated and submitted to 5000 iterations of energy minimization using the PolakeRibiere Conjugated Gradient (PRCG) algorithm [58], AMBER* as force field [59] with the united atoms notation and the implicit model of solvation GB/SA water [60] as implemented in Macro Model ver. 7.2 [61a,b]. For all the studied G-binders the convergence in the conformational search was evaluated using the averaged number of duplicate conformers that resulted always higher than 2, thus indicating a good conformational space exploration. Among all the generated conformations, those within 3 kcal/mol above the energy global minimum were selected and submitted to the SASA analysis. 2.2. BASASA (Bolzmann averaged solvent accessible surface area) descriptor The SASA calculations were computed on defined aromatic portions of the selected compounds. In particular such a descriptor allowed the estimation of the degree of exposition of the planar rings of three different classes of ligands (acridine, perylene and ethidium scaffolds) to the solvent, providing an indirect measure of the stacking potential property of our compounds.

Scheme 2. The BASASA is a weighted descriptor, calculated as the integral of the Boltzmann population (pi) and surface area values of each conformer (SASAi).

S. Alcaro et al. / Biochimie 93 (2011) 1267e1274 Table 1 Chemical structures and IC50 values of compounds 1e18 (Lig). IC50 values, expressed in mM, were measured by TRAP assays. Lig

Chemical structure

1269

Table 1 (continued) Lig

Chemical structure

IC50

Acridine scaffold

N 1

HN

H N

N O

10 0.09

H N

11

N

2

N

N H

O

3 +

0.15

N H

N

N

12

N

O

N H

O

O

N

N

O

O

N

N

N

0.20

N O

N

N

IC50

3.00

N

+

N

O

O

N

N

O

O

H N

13

N+

N H

N

+

N

O

O

N

N

O

O

0.40

2.50

H N

6.80

Ethidium scaffold

N

4

N

N H

O

5

N

N H

N

N

+

N 15

N

O

N H

H N

H2N

13.00

N

N H

+

+

O

N

N N H

O N H

N

NH2 0.019

N

N NH2

N H

0.4

+

N

30.00

N+

N

0.018

NH2 N

16 7

NH2

HN

N H

O N

NH

N

6.00

+

6

N

N

14

O

N H

H N

H2N

5.60

Perylene scaffold

8

9

+

N

N

O

O

N

N

O

O

O

O

N

N

O

O

NH2

H2N +

N

0.03

+

17

N

0.16

18

NH 0.09

NH2

H2N

N

N N

H2N

+

N

0.18

1270

S. Alcaro et al. / Biochimie 93 (2011) 1267e1274

Table 2 Conformational analysis results of compounds 1e18. The most stable conformers (Conf) within 3 kcal/mol above the energy global minimum are reported. Compound 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Rotatable bonds

Conf

11 10 8 8 10 12 14 4 4 12 17 6 12 5 5 3 4 1

3054 386 111 214 387 337 283 10 8 206 334 18 1048 78 79 13 52 2

For all conformers, the SASA, according to the water property suggested by the Macro Model version 9.1 program [62], was performed considering the interaction with a 1.4 Å radius probe. With respect to our previous work related to 9-fluoren derivatives [54], in this analysis we used the graphical user interface, and the SASA value of each conformation was related with respect to its Boltzmann population computed with Scheme (1).

Table 3 Boltzmann Solvent Accessible Surface Area Analysis (BASASA) results for compounds 1e18. Ave, Min and Max indicate, respectively, the averaged, minimum and maximum Solvent Accessible Surface Areas, expressed in Å2. Compound 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Ave

Min

Max

128.91 145.93 140.65 124.32 139.74 111.72 118.27 233.17 225.30 222.29 204.00 160.25 135.77 191.31 188.76 179.83 175.47 198.25

127.88 115.82 124.27 109.21 118.31 93.40 102.31 227.28 225.09 218.27 190.56 158.18 129.23 177.54 188.90 178.62 174.67 197.62

129.63 167.02 142.91 142.14 143.09 126.77 141.99 234.05 225.72 223.18 222.84 160.33 154.94 191.81 196.62 181.23 177.63 198.88

The BASASA descriptor is computed as the integral of Boltzmann weighted SASA values of all conformers, using the population values as coefficients, as reported in Scheme (2). The BASASA is a 1-D descriptor that does not take into account the quality of the interaction between ligand and target. It just gives information related to the available recognition area but nothing

Fig. 1. Energy global minimum conformers of compounds 1e18 (polytube models) obtained in the MC conformational search. Hydrogen bonds are represented with black dashed lines.

S. Alcaro et al. / Biochimie 93 (2011) 1267e1274

Fig. 2. BASASA results for acridine derivatives 1e7. Bar plots refer to the averaged solvent accessible surface area in Å2 for the defined planar aromatic moiety. Error bars regard the maximum and minimum BASASA values in the conformational ensemble generated in the MC search.

about the interaction type. Consequently, BASASA can correlate IC50 within a certain range of values only and it should be used for strictly related structure analogs. In conclusion the BASASA descriptor is a Boltzmann averaged solvent accessible surface area (expressed as Å2) of a defined planar system that takes into account the conformational properties of the ligand. 3. Results In this work we have developed a computational protocol to evaluate a surface area descriptor based on the conformational properties of known G-quadruplex ligands. The molecules have been selected on the basis of their chemical diversity and experimental affinity toward the G-quadruplex human telomeric repeated sequence. We divided the analyzed compounds into three different chemical classes, considering their diverse flexibility and steric hindrance. Specifically, as reported in Table 1, we included in our study 7 acridine (compounds 1e7) [63], 6 perylene (compounds 8e13) [64] and 5 ethidium (compounds 14e18) derivatives [24]. The ability of the selected G-binders to inhibit telomerase activity was reported as telomere repeat amplification protocol (TRAP) assay. In particular, our selection spanned compounds from nano- to micro-molar IC50 activity, in order to verify the discerning reliability of our computational approach. As reported in Table 2, among the three selected classes, acridines resulted as the most flexible compounds; the 3,6,9-trisubstituted 1 generated more than 3000 unique conformations due to the presence of an additional side-chain with respect to the 4,5-disubstituted acridines 2e7. Ethidium derivatives appeared the most rigid among the analyzed G-binders, showing a reduced number of conformations within 3 kcal/mol above the energy global minimum compared to the perylene compounds. In particular, with respect to the lead 18, ethidium derivatives presented a higher conformational variability due to the introduction of additional side chains and/or of substitutions on the free amine groups.

Fig. 3. BASASA results for perylene derivatives 8e13. Bar plots refer to the averaged solvent accessible surface area in Å2 for the defined planar aromatic moiety. Error bars regard the maximum and minimum BASASA values in the conformational ensemble generated in the MC search.

1271

Fig. 4. BASASA results for ethidium derivatives 14e18. Bar plots refer to the averaged solvent accessible surface area in Å2 for the defined planar aromatic moiety. Error bars regard the maximum and minimum BASASA values in the conformational ensemble generated in the MC search.

In the case of perylene ligands, as regards to the reference compound 10, 11 heptacyclic derivative is characterized by the absence of the carbonyl groups in the aromatic core, thus loosing the planarity and increasing its flexibility (Table 2). Moreover the tetracyclic compound 13 presented extended side chains with the highest number of rotatable bonds, thus generating a large set of unique conformations (more than 1000). The side chains of compounds 1e18 were all characterized by hydrogen bond (HB) acceptor and/or donor atoms. The conformation of the side chains was influenced by the established HB network. HB interactions of global minimum energy conformers give an idea of such an influence, but cannot be considered as an exhaustive analysis due to the large amount of conformers found in the MC search (Table 2). As reported in Fig. 1, HB interactions were observed in global minimum conformers of compounds 2, 3, 4, 5, 6, 7 and 13 only, even if the other eleven compounds contained similar HB acceptors and donors. The reason for such a different behavior can be attributed to the nature of the cyclic conformation bridged by the HB. In particular, for acridine derivatives the side chain positions 4 and 5 resulted to be favored for the G-quadruplex binding, while the only possible HB in the perylene class was shown for the most flexible 13. The central core of our theoretical protocol was the BASASA analysis. The analyzed 4,5-dialkylaminoalkylamide acridine ligands 1e7 were divided into three subclasses: the amide tertiary amine substituted derivatives (5 and 6), the amide quaternary ammonium salts (3 and 7) and the 4,5-bis(aminomethyl)-type acridine derivatives 2 and 4. In our analysis, the 3,6,9-trisubstituted acridine 1 was included as reference. As reported in Table 3 and in Fig. 2, within each considered subclass, we obtained the highest BASASA values for the best G-quadruplex stabilizing compounds, with the exception of 1, whose BASASA, equal to 128.91 Å2, was ranked between the maximum and the minimum averaged values. The perylene compounds could also be classified into two main chemical groups: tetracyclic (12 and 13) and heptacyclic (8, 9 and 11) derivatives. Among the heptacyclic ones, other two subclasses were identified: the cyclic amide (8 and 9) and the cyclic amine (11) substituted derivatives. As a reference compound for the perylene scaffold we included the cyclic amide heptacyclic derivative 10. The BASASA analysis, reported in Table 3 and in Fig. 3, presented a wide range of area values comprised from the highest 233.17 Å2 of the most potent 8 to the lowest 135.77 Å2 of the less active 13. In the third considered class, the BASASA results underlined ethidium derivatives 14 and 15 as the most interesting ones (respectively 191.31 Å2 and 188.76 Å2), obtaining the highest area values according to their better binding affinity (Table 3, Fig. 4). These selected compounds were characterized by the presence of a bulky side chain at position 7; by contrast the less active 17 presented the amine groups not-substituted at position 2 and 7, but it reported a diazonium side chain at position 8. Finally the BASASA

1272

S. Alcaro et al. / Biochimie 93 (2011) 1267e1274

values of the reference compound 18 revealed the widest solvent accessible surface area due to the lack of the side chains onto the free above mentioned amines. We have considered the ionization state of the neutral compounds reported in Table 1 taking into account the pH conditions of the TRAP experiments [65]. Repeating the MC and the BASASA simulations we have found no changes into the minimum and maximum values and modest effects into the average data. Since BASASA descriptor could be applied to several different targets, we considered in our analysis also the receptors contribution in order to evaluate the ligands binding affinity. The dynamic equilibrium among the polymorphic conformations of the human monomolecular telomeric sequence is actually well documented [66]. However the predominant pep stacking interactions, responsible to well stabilize G-binders with respect to the G-quartet core, appear to be favored in the G-quadruplex parallel fold (PDB 1KF1) [67]. This observation prompted us to calculate only the SASA of the G-quadruplex X-ray structure, analyzing separately its top and bottom sides (Supplementary Material, Figure S1) and other NMR models (Supplementary Material, Figure S2eS6). 4. Discussion In this analysis the BASASA descriptor provided interesting data taking into account the TRAP assays of the single studied chemical classes of known G-binders. In particular, among the di-substituted acridine compounds, the best G-quadruplex stabilizing 2 resulted to be associated with the highest BASASA value, according to its good binding affinity toward the G-quadruplex human telomeric repeated sequence. A good correlation between the binding affinities of the single studied subclasses and the computed weighted accessible areas was observed, with the exception of the reference most potent 1 ligand. Such a data can be justified by the presence of an additional side chain for compound 1 with respect to the 4,5disubstituted acridines 2-7 likely responsible to increase steric hindrance and consequently reducing the solvent accessibility. Less flexible and more bulky perylene derivatives showed consistent BASASA results, in particular related to the presence of the large heptacyclic scaffold. In fact, it was observed that among 10-related ligands, a conjugated system consisting of more than four rings was required to efficiently target G-quartets. Our approach well reproduced such an observation, that was not in agreement with data reported on G-quadruplex binders structurally related to anthracene or acridine, for which good tetraplex selectivity could be reached in the presence of just three-condensed rings [64]. For the ethidium derivatives 14e17, we have obtained good correlation between their binding affinities and the calculated BASASA values, reproducing their differences in TRAP data. The only exception was noticed for the reference compound 18, that, even showing the lowest G-quadruplex affinity, was associated to the largest accessible surface area. Such a finding was due to the lack of substitutions on the free amine groups at positions 2 and 7, with a resulting increased solvent accessibility. Perylene and ethidium scaffold derivatives BASASA versus IC50 plot analyses are reported in the Supplementary Material respectively in Figure S7 and S8. The BASASA data related to the studied G-binders have to be analyzed taking into account that the area values are related to the whole ligands, with no differentiation between the two possible accessible faces of the molecule, thus taking no notice of the side chains movements and hindrance. By contrast, the SASA evaluation of the G-quadruplex parallel fold presented the advantage to differentiate the contributions of the solvent accessibility derived from the top and the bottom sides of the receptor. The obtained results put in evidence all the studied molecules associated to an overestimated surface accessible area, while the SASA values of

both the receptor sites gave a more realistic idea of the role of the solvent and of the hydrophobicity during the binding event. Such a consideration highlighted the importance of SASA descriptor as a tool to estimate the geometrical complementarity between the telomerase inhibitors and the DNA G-quadruplex target. Moreover our computational approach could rationalize the weight of the ligands aromatic moiety in stacking interactions with respect to the G-tetrad. Hence a broader aromatic surface enhancing stacking interactions with the external G-quartets resulted related to a strong hydrophobic character of the molecule, explaining the Gquadruplex good binding and the telomerase inhibition. With respect to other recent studies [55e57], our current surface area estimation implements the Boltzmann analysis, that takes into account the conformational effects of the ligands. Finally, the BASASA descriptor provided interesting preliminary information to differentiate ligand binding affinities and it could be a useful predictive tool in drug design and optimization processes. However such an application, in this current form, cannot be sufficiently complete to fully rationalize G-binders activity profiles, but it could represent a starting point for a more comprehensive analysis of an extended set of ligands. Furthermore, due to the interesting obtained data, it may be noteworthy to consider other related descriptors, such as p (aromatic), weak polar and aliphatic components of surface area respectively by means of PISA, WPSA and FOSA [68a,b,c] analyses which resulted important for other targets [69,70]. 5. Conclusions In conclusion our studies allowed to rationally explore the conformational space of known G-binders and to analyze their Boltzmann weighted solvent accessible area with respect to their TRAP data. A good correlation between the experimental binding affinities of the single studied classes and the computed weighted accessible areas was observed, indicating such a descriptor as a useful preliminary tool to rationalize molecular activity profiles. Moreover, a future combination of such a geometric analysis with molecular docking studies will be proposed for the rational discovery of novel more active/selective DNA G-quadruplex ligands. The SASA descriptor, applied to ligandereceptor complexes, will be evaluated to improve it for drug design purposes. Acknowledgments This research is supported by the Italian Ministry of Education (Funding for Investments of Base Research) for the years 2009e2014 (code FIRB-IDEAS RBID082ATK_002). Appendix. Supplementary material Supplementary material related to this article can be found online at doi:10.1016/j.biochi.2011.06.014. References [1] R. McElligott, R.J. Wellinger, The terminal DNA structure of mammalian chromosomes, EMBO J. 16 (1997) 3705e3714. [2] C.B. Harley, A.B. Futcher, C.W. Greider, Telomeres shorten during ageing of human fibroblasts, Nature 345 (1990) 458e460. [3] K.A. Olaussen, K. Dubrana, J. Domonta, J.P. Spano, L. Sabatier, J.C. Soria, Telomeres and telomerase as targets for anticancer drug development, Crit. Rev. Oncology/Hematology 57 (2006) 191e214. [4] F. Cuenca, M.J. Moore, K. Johnson, B. Guyen, A. De Cian, S. Neidle, Design, synthesis and evaluation of 4,5-di-substituted acridone ligands with high Gquadruplex affinity and selectivity, together with low toxicity to normal cells, Bioorg. Med. Chem. Lett. 19 (2009) 5109e5113. [5] M. Bisoffi, C.M. Heaphy, J.K. Griffith, Telomeres: prognostic markers for solid tumors, Int. J. Cancer 119 (2006) 2255e2260.

S. Alcaro et al. / Biochimie 93 (2011) 1267e1274 [6] L.R. Kelland, Overcoming the immortality of tumour cells by telomere and telomerase based cancer therapeutics e current status and future prospects, Eur. J. Cancer 41 (2005) 971e979. [7] S. Burge, G.N. Parkinson, P. Hazel, A.K. Todd, S. Neidle, Quadruplex DNA: sequence, topology and structure, Nucleic Acids Res. 34 (2006) 5402e5415. [8] D.J. Patel, A.T. Phan, V. Kuryavyi, Human telomere, oncogenic promoter and 50 -UTR G-quadruplexes: diverse higher order DNA and RNA targets for cancer therapeutics, Nucleic Acids Res. 35 (2007) 7429e7455. [9] A.E. Engelhart, J. Plavec, O. Persil, N.V. Hud (Eds.), Nucleic Acid-Metal Ion Interactions, RSC Publishing, 2008, pp. 118e153. [10] S. Neidle, The structures of quadruplex nucleic acids and their drug complexes, Curr. Opin. Struct. Biol. 19 (2009) 239e250. [11] http://www.rcsb.org. [12] Y. Wang, D.J. Patel, Solution structure of the human telomeric repeat d [AG3(T2AG3)3] G-tetraplex, Structure 1 (1993) 263e282. [13] G.N. Parkinson, M.P. Lee, S. Neidle, Crystal structure of parallel quadruplexes from human telomeric DNA, Nature 417 (2002) 876e880. [14] J.X. Dai, C. Punchihewa, A. Ambrus, D. Chen, R.A. Jones, D. Yang, Structure of the intramolecular human telomeric G-quadruplex in potassium solution: a novel adenine triple formation, Nucleic Acids Res. 35 (2007) 2440e2450. [15] J.X. Dai, M. Carver, C. Punchihewa, R.A. Jones, D. Yang, Structure of the Hybrid2 type intramolecular human telomeric G-quadruplex in Kþ solution: insights into structure polymorphism of the human telomeric sequence, Nucleic Acids Res. 35 (2007) 4927e4940. [16] A.T. Phan, V. Kuryavyi, K.N. Luu, D.J. Patel, Structure of two intramolecular Gquadruplexes formed by natural human telomere sequences in Kþ solution, Nucleic Acids Res. 35 (2007) 6517e6525. [17] D. Sun, B. Thompson, B.E. Cathers, M. Salazar, S.M. Kerwin, J.O. Trent, T.C. Jenkins, S. Neidle, L.H. Hurley, Inhibition of human telomerase by a Gquadruplex-interactive compound, J. Med. Chem. 40 (1997) 2113e2116. [18] A. De Cian, L. Lacroix, C. Dourre, N. Temime-Smaali, C. Trentesaux, J.F. Riou, J.L. Mergny, Targeting telomeres and telomerase, Biochimie 90 (2008) 131e155. [19] (a) A. De Cian, G. Cristofari, P. Reichenbach, E. De Lemos, D. Monchaud, M.P. Teulade-Fichou, K. Shin-Ya, L. Lacroix, J. Lingner, J.L. Mergny, Reevaluation of telomerase inhibition by quadruplex ligands and their mechanisms of action, Proc. Natl. Acad. Sci. U.S.A. 104 (2007) 17347e17352; (b) M. Folini, L. Venturini, G. Cimino-Reale, N. Zaffaroni, Telomeres as targets for anticancer therapies, Expert Opin. Ther. Targets 15 (2011) 579e593. [20] S. Neidle, G. Parkinson, Telomere maintenance as a target for anticancer drug discovery, Nat. Rev. Drug Discov. 1 (2002) 383e393. [21] B. Pagano, C. Giancola, Energetics of quadruplex-drug recognition in anticancer therapy, Curr. Cancer Drug Targets 7 (2007) 520e540. [22] D. Monchaud, M.P. Teulade-Fichou, A hitchhiker’s guide to G-quadruplex ligands, Org. Biomol. Chem. 6 (2008) 627e636. [23] K. Shin-ya, K. Wierzba, K. Matsuo, T. Ohtani, Y. Yamada, K. Furihata, Y. Hayakawa, H. Seto, Telomestatin, a novel telomerase inhibitor from Streptomyces anulatus, J. Am. Chem. Soc. 123 (2001) 1262e1263. [24] F. Koeppel, J.F. Riou, A. Laoui, P. Mailliet, P.B. Arimondo, D. Labit, O. Petigenet, C. Hélène, J.L. Mergny, Ethidium derivatives bind to G-quartets, inhibit telomerase and act as fluorescent probes for quadruplexes, Nucleic Acids Res. 29 (2001) 1087e1096. [25] (a) P.J. Perry, M.A. Read, R.T. Davies, S.M. Gowan, A.P. Reszka, A.A. Wood, L.R. Kelland, S. Neidle, 2,7-Disubstituted amidofluorenone derivatives as inhibitors of human telomerase, J. Med. Chem. 42 (1999) 2679e2684; (b) S. Alcaro, A. Artese, J.N. Iley, S. Missailidis, F. Ortuso, L. Parrotta, R. Pasceri, F. Paduano, C. Sissi, F. Trapasso, M.G. Vigorita, Rational design, synthesis, biophysical and antiproliferative evaluation of fluorenone derivatives with DNA G-quadruplex binding properties, ChemMedChem 5 (2010) 575e583; (c) S. Alcaro, A. Artese, J.N. Iley, R. Maccari, S. Missailidis, F. Ortuso, R. Ottana, P. Ragazzon, M.G. Vigorita, Tetraplex DNA specific ligands based on the fluorenone-carboxamide scaffold, Bioorg. Med. Chem. Lett. 17 (2007) 2509e2514. [26] R.J. Harrison, A.P. Reszka, S.M. Haider, B. Romagnoli, J. Morrell, M.A. Read, S.M. Gowan, C.M. Incles, L.R. Kelland, S. Neidle, Evaluation of by disubstituted acridone derivatives as telomerase inhibitors: the importance of G-quadruplex binding, Bioorg. Med. Chem. Lett. 14 (2004) 5845e5849. [27] R.J. Harrison, S.M. Gowan, L.R. Kelland, S. Neidle, Human telomerase inhibition by substituted acridine derivatives, Bioorg. Med. Chem. Lett. 9 (1999) 2463e2468. [28] O.Y. Fedoroff, M. Salazar, H. Han, V.V. Chemeris, S.M. Kerwin, L.H. Hurley, NMR-Based model of a telomerase-inhibiting compound bound to G-quadruplex DNA, Biochemistry 37 (1998) 12367e12374. [29] (a) W. Duan, A. Rangan, H. Vankayalapati, M.Y. Kim, Q. Zeng, D. Sun, H. Han, O.Y. Fedoroff, D. Nishioka, S.Y. Rha, E. Izbicka, D.D. Von Hoff, L.H. Hurley, Design and synthesis of fluoroquinophenoxazines that interact with human telomeric G-quadruplexes and their biological effects, Mol. Cancer Ther. 1 (2001) 103e120; (b) A.K. Mehta, Y. Shayo, H. Vankayalapati, L.H. Hurley, J. Schaefer, Structure of a quinobenzoxazineeG-quadruplex complex by REDOR NMR, Biochemistry 43 (2004) 11953e11958. [30] P. Alberti, P. Schmidt, C.H. Nguyen, M. Hoarau, D. Grierson, J.L. Mergny, Benzoindoloquinolines interact with DNA tetraplexes and inhibit telomerase, Bioorg. Med. Chem. Lett. 12 (2002) 1071e1074. [31] (a) V. Caprio, B. Guyen, Y. Opoku-Boahen, J. Mann, S.M. Gowan, L.M. Kelland, M.A. Read, S. Neidle, A novel inhibitor of human telomerase derived from

[32]

[33]

[34]

[35]

[36]

[37]

[38]

[39]

[40]

[41]

[42]

[43]

[44] [45]

[46]

[47]

[48]

[49]

[50] [51]

1273

10H-indolo[3,2-b]quinoline, Bioorg. Med. Chem. Lett. 10 (2000) 2063e2066; (b) B. Guyen, C.M. Schultes, P. Hazel, J. Mann, S. Neidle, Synthesis and evaluation of analogues of 10H-indolo[3,2-b]quinoline as G-quadruplex stabilising ligands and potential inhibitors of the enzyme telomerase, Org. Biomol. Chem. 2 (2004) 981e988; (c) L. Guittat, A. Alberti, F. Rosu, S. Van Miert, E. Thetiot, L. Pieters, V. Gabelica, E. De Pauw, A. Ottaviani, J.F. Riou, J.L. Mergny, Interactions of cryptolepine and neocryptolepine with unusual DNA structures, Biochimie 85 (2003) 535e547. J.L. Zhou, Y.J. Lu, T.M. Ou, J.M. Zhou, Z.S. Huang, X.F. Zhu, C.J. Du, X.Z. Bu, L. Ma, L.Q. Gu, Y.M. Li, A.S.C. Chan, Synthesis and evaluation of quindoline derivatives as G-quadruplex inducing and stabilizing ligands and potential inhibitors of telomerase, J. Med. Chem. 48 (2005) 7315e7321. D.F. Shi, R.T. Wheelhouse, D.Y. Sun, L.H. Hurley, Quadruplex-interactive agents as telomerase inhibitors: synthesis of porphyrins and structureactivity relationship for the inhibition of telomerase, J. Med. Chem. 44 (2001) 4509e4523. J.L. Mergny, L. Lacroix, M.P. Teulade-Fichou, C. Hounsou, L. Guittat, M. Hoarau, P.B. Arimondo, J.P. Vigneron, J.M. Lehn, J.F. Riou, T. Garestier, C. Hélène, Telomerase inhibitors based on quadruplex ligands selected by a fluorescence assay, Proc. Natl. Acad. Sci. USA 98 (2001) 3062e3067. J.F. Riou, L. Guittat, P. Mailliet, A. Laoui, E. Renou, O. Petitgenet, F. MégninChanet, C. Hélène, J.L. Mergny, Cell senescence and telomere shortening induced by a new series of specific G-quadruplex DNA ligands, Proc. Natl. Acad. Sci. USA 99 (2002) 2672e2677. C.C. Chang, I.C. Kuo, I.F. Ling, C.T. Chen, H.C. Chen, P.J. Lou, J.J. Lin, T.C. Chang, Detection of quadruplex DNA structures in human telomeres by a fluorescent carbazole derivative, Anal. Chem. 76 (2004) 4490e4494. M. Read, R.J. Harrison, B. Romagnoli, F.A. Tanious, S.M. Gowan, A.P. Reszka, W.D. Wilson, L.R. Kelland, S. Neidle, Structure-based design of selective and potent G quadruplex-mediated telomerase inhibitors, Proc. Natl. Acad. Sci. U.S.A. 98 (2001) 4844e4849. M.J.B. Moore, C.M. Schultes, J. Cuesta, F. Cuanca, M. Gunaratnam, F.A. Tanious, W.D. Wilson, S. Neidle, Trisubstituted acridines as G-quadruplex telomere targeting agents. Effects of extensions of the 3,6- and 9-side chains on quadruplex binding, telomerase activity, and cell proliferation, J. Med. Chem. 49 (2006) 582e599. A.M. Burger, F. Dai, C.M. Schultes, A.P. Reszka, M.J. Moore, J.A. Double, S. Neidle, The G-quadruplex-interactive molecule BRACO-19 inhibits tumor growth, consistent with telomere targeting and interference with telomerase function, Cancer Res. 65 (2005) 1489e1496. C. Martins, M. Gunaratnam, J. Stuart, V. Makwana, O. Greciano, A.P. Reszka, L.R. Kelland, S. Neidle, Structure-based design of benzylamino-acridine compounds as G-quadruplex DNA telomere targeting agents, Bioorg. Med. Chem. Lett. 17 (2007) 2293e2298. C. Pivetta, L. Lucatello, A.P. Krapcho, B. Gatto, M. Palumbo, C. Sissi, Perylene side chains modulate G-quadruplex conformation in biologically relevant DNA sequences, Bioorg. Med. Chem. 16 (2008) 9331e9339. F. Rosu, E. De Pauw, L. Guittat, P. Alberti, L. Lacroix, P. Mailliet, J.F. Riou, J.L. Mergny, Selective interaction of ethidium derivatives with quadruplexes: an equilibrium dialysis and electrospray ionization mass spectrometry analysis, Biochemistry 42 (2003) 10361e10371. J. Kirchmair, S. Distinto, D. Schuster, G. Spitzer, T. Langer, G. Wolber, Enhancing drug discovery through in silico screening: strategies to increase true positives retrieval rates, Curr. Med. Chem. 15 (2008) 2040e2053. J. Verma, V.M. Khedkar, E.C. Coutinho, 3D-QSAR in drug design e a review, Curr. Top. Med. Chem. 10 (2010) 95e115. R.D. Cramer, D.E. Patterson, J.D. Bunce, Comparative molecular field analysis (CoMFA). 1. Effect of shape on binding of steroids to carrier proteins, J. Am. Chem. Soc. 110 (1988) 5959e5967. N. Stiefl, G. Bringmann, C. Rummey, K. Baumann, Evaluation of extended parameter sets for the 3D-QSAR technique MaP: implications for interpretability and model quality exemplified by antimalarially active naphthylisoquinoline alkaloids, J. Comput.Aided. Mol. Des 17 (2003) 347e365. B.D. Silverman, D.E. Platt, Comparative molecular moment analysis (CoMMA): 3D-QSAR without molecular superposition, J. Med. Chem. 39 (1996) 2129e2140. R. Bursi, T. Dao, T. van Wijk, M. de Gooyer, E. Kellenbach, P. Verwer, Comparative spectra analysis (CoSA): spectra as three-dimensional molecular descriptors for the prediction of biological activities, J. Chem. Inf. Comput. Sci. 39 (1999) 861e867. (a) B. Lee, F.M. Richards, The interpretation of protein structures: estimation of static accessibility, J. Mol. Biol. 55 (1971) 379e400; (b) S. Sprang, D. Yang, R.J. Fletterick, Solvent accessibility properties of complex proteins, Nature 280 (1979) 333e335; (c) V.J. Hilser, J. Gómez, E. Freire, The enthalpy change in protein folding and binding: refinement of parameters for structure-based calculations, Proteins 26 (1996) 123e133; (d) R.S. Spolar, J.H. Ha, M.T. Record, Hydrophobic effect in protein folding and other noncovalent processes involving proteins, Proc. Natl. Acad. Sci. U S A 86 (1989) 8382e8385. Y. Ofran, V. Mysore, B. Rost, Prediction of DNA-binding residues from sequence, Bioinformatics 23 (2007) i347ei353. A.A. Lipanov, D.B. Beglov, V.P. Chuprina, DNA B to D transition can be explained in terms of hydration economy of the minor groove atoms, J. Mol. Biol. 210 (1989) 399e409.

1274

S. Alcaro et al. / Biochimie 93 (2011) 1267e1274

[52] S. Ahmad, Sequence-dependence and prediction of nucleotide solvent accessibility in double stranded DNA, Gene 428 (2009) 25e30. [53] I. Haq, J.E. Ladbury, B.Z. Chowdhry, T.C. Jenkins, J.B. Chaires, Specific binding of hoechst 33258 to the d(CGCAAATTTGCG)2 duplex: calorimetric and spectroscopic studies, J. Mol. Biol. 271 (1997) 244e257. [54] S. Alcaro, A. Arena, S. Neri, R. Ottanà, F. Ortuso, B. Pavone, M.G. Vigorita, Design and synthesis of DNA-intercalating 9-fluoren-b-O-glycosides as potential IFN-inducers, and antiviral and cytostatic agents, Bioorg. Med. Chem. 12 (2004) 1781e1791. [55] S. Agrawal, R.P. Ojha, S. Maiti, Energetics of the human Tel-22 quadruplextelomestatin interaction: a molecular dynamics study, J. Phys. Chem. B. 112 (2008) 6828e6836. [56] L. Aixiao, M. François, B. Florent, D. Michel, W. Baoshan, Z. Xiang, W. Ping, Molecular modeling study of binding site selectivity of TQMP to G-quadruplex DNA, Eur. J. Med. Chem. 45 (2010) 983e991. [57] W. Xu, J.H. Tan, S.B. Chen, J.Q. Hou, D. Li, Z.S. Huang, L.Q. Gu, Studies on the binding of 5-N-methylated quindoline derivative to human telomeric G-quadruplex, Biochem. Biophys. Res. Commun. 406 (2011) 454e458. [58] E. Polak, G. Ribiére, Note sur la convergence de directions conjuguées, Rev. Francaise Infomat Recherche Operatonelle, 3e Année, vol. 16 (1969) pp. 35e43. [59] D.Q. McDonald, W.C. Still, AMBER torsional parameters for the peptide backbone, Tetrahedron Lett. 33 (1992) 7743e7746. [60] W.C. Still, A. Tempczyk, R.C. Hawley, T. Hendrickson, Semianalytical treatment of solvation for molecular mechanics and dynamics, J. Am. Chem. Soc. 112 (1990) 6127e6129. [61] (a) Maestro v. 4.1. Schrödinger Inc., Portland, OR (USA), 1998e2001; (b) F. Mohamadi, N.G.J. Richards, W.C. Guida, R. Liskamp, M. Lipton, C. Caufield, G. Chang, T. Hendrickson, W.C. Still, Macromodel-an integrated

[62] [63]

[64]

[65] [66] [67]

[68]

[69]

[70]

software system for modeling organic and bioorganic molecules using molecular mechanics, J. Comput. Chem. 11 (1990) 440e467. Maestro v. 9.1. Schrödinger Inc., Portland, OR (USA), 2010. M. Laronze-Cochard, Y.M. Kim, B. Brassart, J.F. Riou, J.Y. Laronze, J. Sapi, Synthesis and biological evaluation of novel 4,5-bis(dialkylaminoalkyl)substituted acridines as potent telomeric G-quadruplex ligands, Eur. J. Med. Chem. 44 (2009) 3880e3888. C. Sissi, L. Lucatello, A.P. Krapcho, D.J. Maloney, M.B. Boxer, M.V. Camarasa, G. Pezzoni, E. Menta, M. Palumbo, Tri-, tetra- and heptacyclic perylene analogues as new potential antineoplastic agents based on DNA telomerase inhibition, Bioorg. Med. Chem. 15 (2007) 555e562. LigPrep 2.3 Schrödinger Suite. Schrödinger, LLC, New York, NY, USA, 2009. A.N. Lane, J.B. Chaires, R.D. Gray, J.O. Trent, Stability and kinetics of G-quadruplex structures, Nucleic Acids Res. 36 (2008) 5482e5515. S.E. Evans, M.A. Mendez, K.B. Turner, L.R. Keating, R.T. Grimes, S. Melchoir, V.A. Szalai, End-stacking of copper cationic porphyrins on parallel-stranded guanine quadruplexes, J. Biol. Inorg. Chem. 12 (2007) 1235e1249. (a) Schrödinger Suite. Schrödinger, LLC, New York, NY, USA, 2009; (b) W.L. Jorgensen, J. Chandrasekhar, J.D. Madura, R.W. Impey, M.L. Klein, Comparison of simple potential functions for simulating liquid water, J. Chem. Phys. 79 (1983) 926e935; (c) W.L. Jorgensen, E.M. Duffy, Prediction of drug solubility from structure, Adv. Drug Deliv. Rev. 54 (2002) 355e366. S. Umamatheswari, B. Balaji, M. Ramanathan, S. Kabilan, Synthesis, antimicrobial evaluation and QSAR studies of novel piperidin-4-yl-5-spiro-thiadiazoline derivatives, Bioorg. Med. Chem. Lett. 20 (2010) 6909e6914. S. Nunez, J. Venhorst, C.G. Kruse, Assessment of a novel scoring method based on solvent accessible surface area descriptors, J. Chem. Inf. Model. 50 (2010) 480e486.