Inhibitor Selectivity for Cyclin-Dependent Kinase†7: A†Structural, Thermodynamic, and Modelling Study.

Deregulation of the cell cycle by mechanisms that lead to elevated activities of cyclin-dependent kinases (CDK) is a feature of many human diseases, cancer in particular. We identified small-molecule inhibitors that selectively inhibit CDK7, the kinase that phosphorylates cell-cycle CDKs to promote their activities. To investigate the selectivity of these inhibitors we used a combination of structural, biophysical, and modelling approaches. We determined the crystal structures of the CDK7-selective compounds ICEC0942 and ICEC0943 bound to CDK2, and used these to build models of inhibitor binding to CDK7. Molecular dynamics (MD) simulations of inhibitors bound to CDK2 and CDK7 generated possible models of inhibitor binding. To experimentally validate these models, we gathered isothermal titration calorimetry (ITC) binding data for recombinant wild-type and binding site mutants of CDK7 and CDK2. We identified specific residues of CDK7, notably Asp155, that are involved in determining inhibitor selectivity. Our MD simulations also show that the flexibility of the G-rich and activation loops of CDK7 is likely an important determinant of inhibitor specificity similar to CDK2.

A novel pyrazolo[1,5-a]pyrimidine is a potent inhibitor of cyclin-dependent protein kinases 1, 2, and 9, which demonstrates antitumor effects in human tumor xenografts following oral administration.

Cyclin-dependent protein kinases (CDKs) are central to the appropriate regulation of cell proliferation, apoptosis, and gene expression. Abnormalities in CDK activity and regulation are common features of cancer, making CDK family members attractive targets for the development of anticancer drugs. Here, we report the identification of a pyrazolo[1,5-a]pyrimidine derived compound, 4k (BS-194), as a selective and potent CDK inhibitor, which inhibits CDK2, CDK1, CDK5, CDK7, and CDK9 (IC50= 3, 30, 30, 250, and 90 nmol/L, respectively). Cell-based studies showed inhibition of the phosphorylation of CDK substrates, Rb and the RNA polymerase II C-terminal domain, down-regulation of cyclins A, E, and D1, and cell cycle block in the S and G2/M phases. Consistent with these findings, 4k demonstrated potent antiproliferative activity in 60 cancer cell lines tested (mean GI50= 280 nmol/L). Pharmacokinetic studies showed that 4k is orally bioavailable, with an elimination half-life of 178 min following oral dosing in mice. When administered at a concentration of 25 mg/kg orally, 4k inhibited human tumor xenografts and suppressed CDK substrate phosphorylation. These findings identify 4k as a novel, potent CDK selective inhibitor with potential for oral delivery in cancer patients.

Quadruplex DNA: sequence, topology and structure.

G-quadruplexes are higher-order DNA and RNA structures formed from G-rich sequences that are built around tetrads of hydrogen-bonded guanine bases. Potential quadruplex sequences have been identified in G-rich eukaryotic telomeres, and more recently in non-telomeric genomic DNA, e.g. in nuclease-hypersensitive promoter regions. The natural role and biological validation of these structures is starting to be explored, and there is particular interest in them as targets for therapeutic intervention. This survey focuses on the folding and structural features on quadruplexes formed from telomeric and non-telomeric DNA sequences, and examines fundamental aspects of topology and the emerging relationships with sequence. Emphasis is placed on information from the high-resolution methods of X-ray crystallography and NMR, and their scope and current limitations are discussed. Such information, together with biological insights, will be important for the discovery of drugs targeting quadruplexes from particular genes.

Topology variation and loop structural homology in crystal and simulated structures of a bimolecular DNA quadruplex.

The topology of DNA quadruplexes depends on the nature and number of the nucleotides linking G-quartet motifs. To assess the effects of a three-nucleotide TTT linker, the crystal structure of the DNA sequence d(G(4)T(3)G(4)) has been determined at 1.5 A resolution, together with that of the brominated analogue d(G(4)(Br)UTTG(4)) at 2.4 A resolution. Both sequences form bimolecular intermolecular G-quadruplexes with lateral loops. d(G(4)(Br)UTTG(4)) crystallized in the monoclinic space group P2(1) with three quadruplex molecules in the asymmetric unit, two associating together as a head-to-head stacked dimer, and the third as a single head-to-tail dimer. The head-to-head dimers have two lateral loops on the same G-quadruplex face and form an eight-G-quartet stack, with a linear array of seven K(+) ions between the quartets. d(G(4)T(3)G(4)) crystallized in the orthorhombic space group C222 and has a structure very similar to the head-to-tail dimer in the P2(1) unit cell. The sequence studied here is able to form several different folds; however, all four quadruplexes in the two structures have lateral loops, in contrast to the diagonal loops reported for the analogous quadruplex with T(4) loops. A total of seven independent T(3) loops were observed in the two structures. These can be classified into two discrete conformational classes, suggesting that these represent preferred loop conformations that are independent of crystal-packing forces.

Predictive modelling of topology and loop variations in dimeric DNA quadruplex structures.

We have used a combination of simulated annealing (SA), molecular dynamics (MD) and locally enhanced sampling (LES) methods in order to predict the favourable topologies and loop conformations of dimeric DNA quadruplexes with T2 or T3 loops. This follows on from our previous MD simulation studies on the influence of loop lengths on the topology of intramolecular quadruplex structures [P. Hazel et al. (2004) J. Am. Chem. Soc., 126, 16 405-16 415], which provided results consistent with biophysical data. The recent crystal structures of d(G4T3G4)2 and d(G4BrUT2G4) (P. Hazel et al. (2006) J. Am. Chem. Soc., in press) and the NMR-determined topology of d(TG4T2G4T)2 [A.T. Phan et al. (2004) J. Mol. Biol., 338, 93-102] have been used in the present study for comparison with simulation results. These together with MM-PBSA free-energy calculations indicate that lateral T3 loops are favoured over diagonal loops, in accordance with the experimental structures; however, distinct loop conformations have been predicted to be favoured compared to those found experimentally. Several lateral and diagonal loop conformations have been found to be similar in energy. The simulations suggest an explanation for the distinct patterns of observed dimer topology for sequences with T3 and T2 loops, which depend on the loop lengths, rather than only on G-quartet stability.

Loop-length-dependent folding of G-quadruplexes.

Guanine-rich DNA sequences can form a large number of structurally diverse quadruplexes. These vary in terms of strand polarity, loop composition, and conformation. We have derived guidelines for understanding the influence of loop length on the structure adopted by intramolecular quadruplex-forming sequences, using a combination of experimental (using CD and UV melting data) and molecular modeling and simulation techniques. We find that a parallel-stranded intramolecular quadruplex structure is the only possible fold when three single residue loops are present. When single thymine loops are present in combination with longer length loops, or when all loops are longer than two residues, both parallel- and antiparallel-folded structures are able to form. Multiple conformations of each structure are likely to coexist in solution, as they were calculated to have very similar free energies.

Synthesis and evaluation of analogues of 10H-indolo[3,2-b]quinoline as G-quadruplex stabilising ligands and potential inhibitors of the enzyme telomerase.

We report here the synthesis and evaluation for telomerase-inhibitory and quadruplex DNA binding properties of several rationally-designed quindoline analogues, substituted at the 2- and 7- positions. The ability of these compounds to interact with and stabilise an intramolecular G-quadruplex DNA against increases in temperature was evaluated by a fluorescence-based (FRET) melting assay. The resulting T(m) values were found to correlate with their potency for telomerase inhibition, as measured in an in vitro telomerase TRAP assay. The interactions of a number of compounds with a quadruplex DNA molecular structure were simulated by molecular modelling methods. It is concluded that this class of compound represents a new chemical type suitable for further development as telomerase inhibitors.