Impact of kinesin Eg5 inhibition by 3,4-dihydropyrimidin-2(1H)-one derivatives on various breast cancer cell features

Abstract

Background: Breast cancer is a complex heterogeneous disease and is one of the leading causes of death among women. In addressing the need for treatments of this life-threatening illness, we studied 3,4-dihydropyrimidin-2(1H)-one (or thione) derivatives (DHPMs), a class of inhibitor molecules of the Eg5 motor spindle protein that shows pronounced antitumor activity against several cancer cell lines.

Methods: An in vitro screening was performed for identification of DHPMs with potent antitumor effects on MCF-7 and MDA-MB-231 cells and the selected DHPMs were evaluated for their inhibitory activity on Eg5 both in silico, using Molecular dynamics, and in vitro Eg5 inhibition assays. Analysis of cell death induction, proliferation, cell cycle and cancer stem cells (CSC) profile were performed by flow cytometry to assess the influence of the selected DPHMs on these important tumor features. Finally, the effects of DHPM treatment on tube formation were evaluated in vitro using HUVEC cells, and in vivo using a model on chorioallantoic membrane (CAM) of fertilized eggs.

Results: We identified five DHPMs with pronounced inhibitory activity on Eg5 motor protein interfering with the proper mitotic spindle assembly during cell division. These compounds impair the correct conclusion of cell cycle of the breast cancer cells and showed to be selective for tumor cells. Moreover, DHPMs modulate the CD44(+)/CD24(-) phenotype leading to a decrease in the CSC population in MDA-MB-231 cells, an important effect since CSC are resistant to many conventional cancer therapies and play a pivotal role in tumor initiation and maintenance. This observation was confirmed by the results which demonstrated that DHPM treated cells had impaired proliferation and were unable to sustain angiogenesis events. Finally, the DHMP treated cells were induced to apoptosis, which is one of the most pursued goals in drug development.

Conclusions: The results of our study strongly suggest that DHPMs inhibit important tumorigenic features of breast cancer cells leading them to death by apoptosis. These findings firmly point to DHPM molecular architecture as a promising alternative against breast cancer.

Figure 1

DHPM derivatives produce a dose and time dependent cytotoxicity and are selective for tumor cells. A, MCF-7 and MDA-MB-231 cells were treated for 72 h with 4 m, 4bt (dimethylenastron), 4p, 4bc and 4x (100 μM – 1 mM) and cell viability evaluated by the MTT assay. Columns, mean of viable cells; bars, SEM; ***(P < 0.001). B, Fibroblasts were treated with the compounds and concentrations that showed significant activities in breast tumor cells at 72 h and the results compared with those from MCF-7 and MDA-MB-231 cells. Columns, mean of viable cells; bars, SEM; **P < 0.01 and ***P < 0.001 compared with treated fibroblasts. C, MCF-7 and MDA-MB-231 cells were treated with 4 m, 4bt (dimethylenastron), 4p, 4bc and 4x at the maximum non-cytotoxic concentration to normal cells for 24 h, 48 h and 72 h and cell viability evaluated in function of time by the MTT assay. Columns, mean of viable cells; bars, SEM. A, B, C, Data represent the mean ± SEM of three independent experiments in triplicates.

Figure 2

DHPM derivatives cause severe morphological alterations on breast cancer cells. MCF-7 and MDA-MB-231 cells were treated with the maximum non-cytotoxic concentration to normal cells for 72 h and morphological alterations analyzed under an inverted light microscope. Cells with elongated morphology, adherent to the surface and normal size are shown by arrows. Cells with size reduction, absence of focal adhesion points, rounded morphology and detached from the surface where they had been seeded are shown by arrowheads.

Figure 3

3D structure-based models of the Eg5 complexes after the MD assays. Molecular dynamics simulations (MD) of the Eg5 protein and Eg5 protein complexes with each of the five tested compounds were conducted to assess the DHMPs influence on Eg5 movements. The panels show ribbon representations of the Eg5 and Licorice representations of the tested compounds and monastrol. The yellow dotted lines show the polar contacts responsible for the maintenance of molecule linkage with the protein binding-site. The binding-region is perfectly conserved, but not the binding-residues, which lead us to realize the influence of the compound topology. Major differences may be seen in the interactions between 4 m, 4x and 4p and the Eg5 binding-site, which show completely different binding-residues. However, 4bt and 4bc retain the linkage with the Arg119 of the Eg5 when comparing to the monastrol, showing its importance in the inhibition processes. Indirect links mediated by water molecules (cyan), occur in residues (green) annotated in orange; direct links occur with residues annotated in black.

Figure 4

4bc leads Eg5 to a very restricted conformation and potently inhibits this protein in vitro.A, Essential dynamics analysis over time was performed to assess the DHPMs interference on Eg5 protein stability. Values far from zero mean that the structure has greater movement amplitude in its first principal component. It is possible to note that compound 4bc leads the protein to a more stable conformation after the MD assays. B, RMSD from the initial structure over time. Compound 4bc allows the greatest initial shift compared to the other molecules, but all of them lead the protein to a stable conformation after about 15 ns. C, Inhibitory rate of Eg5 by DHPMs in vitro. Kinesin Eg5 (1 μg) was incubated with the IC₅₀ of 4 m, 4bt (dimethylenastron), 4p, 4bc, 4x and monastrol with readings taken immediately after incubation at room temperature at 30 second intervals for a total reaction time of 30 minutes. Reactions were measured in spectrophotometer set in kinetic mode and an absorbance wavelength of 360 nm. Dots, mean of nmols of phosphate versus reaction time. *P < 0.05.

Figure 5

DHPMs produce monoastral spindles and 4p can lead cancer cells to a mitotic catastrophe phenotype. A, MCF-7 cells exposed to 4 m (1.0 mM), 4bt (dimethylenastron, 0.8 mM), 4p (0.4 mM), 4bc (1.0 mM), 4x (0.8 mM), monastrol (0.1 mM – positive control) or to culture medium only (negative control) for 24 h were immunostained with α-tubulin (green) and DNA stained with DAPI (blue). Cells were analyzed under a laser scanning confocal microscope. B, MCF-7 cells control or treated with 4p (0.4 mM) for 24 h had their nucleus stained with DAPI (left panels) or were processed and analyzed by Transmission Electron Microscopy after exposure to 4p (0.4 mM) for 48 h (right panels). Treated cells (lower panels) showed nuclear fragmentation with multiple micronuclei formation (MN), perinuclear vacuolization (V) and maintenance of cell membrane integrity (black arrows) compared to untreated cells (upper panels) showed nuclei with normal morphology and integrity (N) and (white arrowhead).

Figure 6

DHPMs potently repress breast tumor cell proliferation, induce apoptosis and modulate the CSC phenotype. MCF-7 and MDA-MB-231 cells control or cells treated with 4 m (1.0 mM), 4bt (dimethylenastron 0.8 mM), 4p (0.4 mM), 4bc (1.0 mM), 4x (0.8 mM) and monastrol (1 mM) were analyzed by flow cytometry to assess: the type of cell death caused by treatment for 72 h by staining cells with Annexin-V and PI (A); the ability to inhibit cell proliferation after 72 h of DHPM exposure, using CellTrace™ CFSE (B); the cell cycle profile after 48 h of treatment by staining the DNA content with PI (C); and the CSC population after 24 h of treatment, evaluating the CD44 and CD24 expression (D). Columns, mean of cells; bars, SEM, *P < 0.05; **P < 0.01; ***P < 0.001. All analyzes were performed in three independent experiments.

Figure 7

DHPM derivatives inhibit angiogenesis both in vitro and in vivo.A, Tube formation analysis by HUVEC cells seeded onto the surface of the polymerized ECMatrix™ in 96-well plates after 11 h of treatment with 4 m, 4bt (dimethylenastron), 4p, 4bc, 4x and monastrol. Representative images of cells treated with 300 μM of each compound documented using an inverted light microscope at 20x magnification (left panel). Tube-like structures formed by HUVEC cells control (asterisks) or cells exposed to 30 μM, IC₅₀ or 300 μM of 4 m, 4bt (dimethylenastron), 4p, 4bc, 4x and monastrol for 11 h, which have lost their ability to cluster in tube-like structures (arrows), were quantified by pattern recognition (right panel). Columns, means of scored tube formation; bars, SEM; **P < 0.01; ***P < 0.001 compared with untreated control. Seven frames per group of three independent experiments (n = 1) were quantified. B, Chorioallantoic membrane (CAM) tumor assay. MCF-7 and MDA-MB-231 cells control or treated with 500 ng of 4 m, 4bt (dimethylenastron), 4p, 4bc, 4x, monastrol or genistein (antiangiogenic control) were implanted into CAM and on the 17th day of development the quantification of total nets was performed by Wimasis Image Analysis (left panel). Columns, mean of total nets (n = 3–5); bars, SEM. A representative image of the most effective treatment for each cell compared with the respective controls can be observed (right panels).