Combinatorial Chemistry & High Throughput Screening - Volume 17, Issue 10, 2014

Acridinone derivatives as imidazoacridinones and triazoloacridinones are the new potent antitumor agents characterized by different mechanisms of action related to their ability to interact with DNA. The analysis undertaken in this study involves searching of QSAR (Quantitative Structure-Activity Relationship) and QSRR (Quantitative Structure- Retention Relationship) models, which would allow to predict the biological activity of acridinones expressed as the ability to stabilize the secondary structure of DNA (ΔT), based on their structural parameters and chromatographic retention data. For this purpose, 20 acridinone derivatives were subjected to chromatographic analyses and molecular modeling, followed by statistical analyses using multiple linear regression method (MLR). As a novelty aspect, except for RP-HPLC approach, hydrophilic interaction chromatography (HILIC) columns were tested. As a result of performed analysis, appropriate QSAR and QSRR models were obtained, and each model was analyzed in terms of prediction of acridinones’ ability to interact with DNA. Derived QSAR and QSRR models were characterized as one, with good prediction performance. Conclusively, the proposed connected QSAR and QSRR strategies allow to predict in silico the ability of acridinones to interact with DNA without the necessity of performing any biological experiments under in vitro and in vivo conditions.

A novel hollow fiber cell fishing with high performance liquid chromatography (HFCF-HPLC) was extended and used to screen flavonoid and anthraquinone active compound groups simultaneously from traditional Chinese medicines (TCMs). In this study, three cells (MCF-7, SGC7901, and MADB-106) were seeded on the inner wall of the hollow fiber employed to screen bioactive components from TCM water decoction. The variables influencing HFCFHPLC, such as cell seeding time, screening stirring rate and time, and active compound concentration, were investigated and optimized. The surface property of the hollow fiber seeded with cells, the cell survival rate under different conditions, the nonspecific binding between active centers in the fiber and the target compounds, and the repeatability and recovery of HFCF-HPLC were analyzed and validated. Certain structures of the compounds fished by HFCF-HPLC were identified after comparing the retention times of the reference substances. To verify preliminarily the binding site between the bioactive components and cells, we separated the cell membrane and cell organelle from live MCF-7 cells. We then employed the cell membrane, cell organelle, and the whole cells to screen simultaneously the active compounds. The cell fishing factor of the active compound was calculated and discussed as the index of cell-drug binding ability in HFCFHPLC. Tamoxifen as a positive control and indomethacin as a negative control were screened by HFCF-HPLC to verify the method. The results indicate that HFCF-HPLC is an effective and reliable method for the screening and analysis of bioactive components. Moreover, this method can be applied to predict bioactive candidates in TCMs.

The cell division cycle 25 (Cdc25) family of proteins is a group of highly conserved dual specificity phosphatases that regulate cyclin-dependent kinases and represent a group of attractive drug targets for anticancer therapies. To develop novel Cdc25B inhibitors, the ZINC database was screened for finding optimal fragments with de novo design approaches. As a result, top 11 compounds with higher binding affinities in flexible docking were obtained, which were derived from five novel scaffolds (scaffold C) consisting of the linker-part and two isolated scaffolds (scaffold A and B)located in the two binding domains (catalytic pocket and swimming pool), respectively. The subsequent molecular docking and molecular dynamics studies showed that these compounds not only adopt more favorable conformations but also have stronger binding interaction with receptor than the inhibitors identified previously. The additional absorption, distribution, metabolism, excretion and toxicity (ADMET) predictions also indicted that the 11 compounds (especially Comp#1) hold a high potential to be novel lead compounds for anticarcinogen. Consequently, the findings reported here may at least provide a new strategy or useful insights for designing effective Cdc25B inhibitors.

Polo-like kinase 1 (Plk1) is a decisive enzyme with its multifunctional activity in cell cycle progression especially mitosis. The over expression of Plk1 in broad spectrum of cancer types turns into a promising therapeutic target against cancer. In the present study, several ligand based pharmacophore models and atom based 3D-QSAR models have been generated using a series of 49 thiazole and thiophene derivatives with well prescribed Plk1 inhibitory activity. From the generated models, the AADRR hypothesis associated with an atom based 3D-QSAR model provided a satisfactory statistical significance containing predictive ability of 40 training set (R2 = 0.9539, SD = 0.1789, F = 113.8) and nine test set (Q2 = 0.4868, RMSE = 0.5333, Pearson R = 0.7114) molecules. The Hypothesis, AADRR explains the salient structural features of the molecules and the developed 3D-QSAR model points out the effect of hydrophobic groups, electron withdrawing groups and H-bond donor groups on Plk1 inhibition for the most active compound 47. The results were further supported by molecular docking studies, which explain the hydrogen bond interactions and binding mode of the ligands with Plk1. These molecular modelling results are expected to be useful for further design of active Plk1 inhibitors.

Ovarian cancer is among the leading causes of cancer deaths in women, and is the most fatal gynecological malignancy. Poor outcomes of the disease are a direct result of inadequate detection and diagnostic methods, which may be overcome by the development of novel efficacious screening modalities. However, the advancement of such technologies is often time-consuming and costly. To overcome this hurdle, our laboratory has established a time and cost effective method of selecting and identifying ovarian carcinoma avid bacteriophage (phage) clones using high throughput phage display technology. These phage clones were selected from a filamentous phage fusion vector (fUSE5) 15-amino acid peptide library against human ovarian carcinoma (SKOV-3) cells, and identified by DNA sequencing. Two phage clones, pM6 and pM9, were shown to exhibit high binding affinity and specificity for SKOV-3 cells using micropanning, cell binding and fluorescent microscopy studies. To validate that the binding was mediated by the phage-displayed peptides, biotinylated peptides (M6 and M9) were synthesized and the specificity for ovarian carcinoma cells was analyzed. These results showed that M6 and M9 bound to SKOV-3 cells in a dose-response manner and exhibited EC50 values of 22.9 ± 2.0 μM and 12.2 ± 2.1μM (mean ± STD), respectively. Based on this, phage clones pM6 and pM9 were labeled with the near-infrared fluorophore AF680, and examined for their pharmacokinetic properties and tumor imaging abilities in vivo. Both phage successfully targeted and imaged SKOV-3 tumors in xenografted nude mice, demonstrating the ability of this method to quickly and cost effectively develop novel ovarian carcinoma avid phage

Nicotinamide phosphoribosyltransferase (NAMPT), an enzyme taking part in main NAD biosynthetic pathway, is an attractive target for anticancer therapy. The purpose of our study is to find novel NAMPT inhibitors based on in silico drug discovery means including the generation of 3D-QSAR models, and virtual screening techniques. Firstly, ten pharmacophore models were generated by Catalyst/HypoGen algorithm. Hypo1 with high correl value (0.96), large Δcost (77.77), and low root mean square deviation (0.81), featured by four chemical features was selected as the best one. Subsequently, Hypo1 was validated through test set prediction and Fischer’s randomization methodologies. Then we screened some public compound libraries (Asinex, Ibscreen and Natural products database) using Hypo1 for a 3D query. The screened hits were further refined by Lipinski’s rule of five, ADMET properties as well as molecular docking studies. Finally, six molecules with diverse scaffolds exhibited the right pharmacophore features and good binding modes between the receptor and ligands, and were selected as possible candidates against NAMPT for further study.

A non-linear quantitative structure activity relationship (QSAR) model based on 350 drug molecules was developed as a predictive tool for drug protein binding, by correlating experimentally measured protein binding values with ten calculated molecular descriptors using a radial basis function (RBF) neural network. The developed model has a statistically significant overall correlation value (r > 0.73), a high efficiency ratio (0.986), and a good predictive squared correlation coefficient (q2) of 0.532, which is regarded as producing a robust and high quality QSAR model. The developed model may be used for the screening of drug candidate molecules that have high protein binding data, filtering out compounds that are unlikely to be protein bound, and may assist in the dose adjustment for drugs that are highly protein bound. The advantage of using such a model is that the percentage of a potential drug candidate that is protein bound (PB (%)) can be simply predicted from its molecular structure.

Zinc plays a vital role in structural organization, regulation of function and stabilization of the folded protein, which ultimately activates or inactivates the binding sites of the protein. Its transition makes a major change in the protein and its binding affinity. The ligand binding aggrecanases can be influenced by Zn2+ ions; therefore the study focuses on checking the binding mode in the presence and absence of zinc using Docking and Molecular dynamics simulation. The crystal structure with zinc was considered as wild type (ADAMTS-4-1Zn2+, ADAMTS-5-1Zn2+) and the crystal structure without zinc was considered as the mutant type (ADAMTS-4-0Zn2+, ADAMTS-5-0Zn2+). Mutations were made manually by deleting the zinc atom. ADAMTS-4-1Zn2+ had the best Glide score of -12.66 kcal·mol−1, whereas ADAMTS-4-0Zn2+ had -11.69 kcal·mol−1. ADAMTS-4-1Zn2+ had the best glide energy of -72.29 kcal·mol−1, whereas ADAMTS-4-0Zn2+ had -68.44 kcal·mol−1. ADAMTS-4-1Zn2+ had the best glide e-model of -116.34, whereas ADAMTS-4-0Zn2+ had -104.264. The RMSD value for ADAMTS-4-1Zn2+ and ADAMTS-4-0Zn2+ was 1.9. These results suggested that the absence of zinc decreases the binding affinity of ADAMTS-4 with its inhibitor. ADAMTS-5-1Zn2+ had the best Glide score of -8.32 kcal·mol−1, whereas ADAMTS-5-0Zn2+ had -6.62 kcal·mol−1. ADAMTS-5-1Zn2+ had the best glide energy of -70.28 kcal·mol−1, whereas ADAMTS-5-0Zn2+ had -66.02 kcal·mol−1. ADAMTS-5-1Zn2+ had the best glide e-model of -108.484, whereas ADAMTS-5-0Zn2+ had -93.81. The RMSD value for ADAMTS-5-1Zn2+ and ADAMTS-5-0Zn2+ was 0.48Å. These results confirmed that the absence of zinc decreased the binding affinity of ADAMTS-5 with its inhibitor whereas the presence extended the docking energy range and strengthened the binding affinity. Per-residue interaction study, MM-GBSA and Molecular Dynamics showed that all the four complexes underwent extensive structural changes whereas the complex with zinc was stable throughout the simulation period.