Fig 1.
Workflow of the screening strategy.
The graphic shows the number of compounds (triangle on the left), their characteristics (mid row) and the respective approach (right) that was applied for compound selection. Arrows from top to bottom indicate the “timeline“. *described in [19].
Fig 2.
Compound selection process and compound chemical structures.
A) Scatter plot shows distribution of 25,000 compounds in regard to cell viability (%) in U2OS cell line with compounds not significantly affecting viability (grey dots), those that reduced viability from 75 to 20% (blue dots), and defined active compounds that were selected for further analysis with viability ranging from 20 to 65% (green dots). Cell viability was determined using Celltiter Blue (CtB) assay. B) Hierarchical clustering of 320 defined active compounds. Compounds were screened on various OS and control cell lines. Respective cell lines are indicated on the left, color key shows values according to cell viability ranging from 20–100%. The red circle marks the cluster that was short-listed for subsequent medicinal chemistry analysis. C) The graph shows inter-replicate variability in the primary and the counter screen of the 29 short-listed compounds. A low inter-replicate variability was used as a selection criteria in medicinal chemistry analysis. The two confirmed hits are highlighted in red and blue, respectively. D) and E) show structures of the two confirmed OS-selective hits, namely compounds A13 (D) and H12 (E).
Table 1.
Cell type characteristics and proliferation potential.
Fig 3.
Dose-response analyses in OS cell lines and control cells after compound treatment.
Cells were treated for 24h with compounds A13 (A), H12 (B), or with classical drugs staurosporine (C) and doxorubicin (D), respectively. Cell viability was assessed using CtB assay. Shown is fold change of cell viability normalized to DMSO-treated control for each cell type after incubation with increasing concentrations (μM) of the respective substance for 24h. Data are expressed as mean +/- SD from duplicates of three independent experiments. Differences of means were calculated using multiple t test for all cell lines versus primary hOB and all highly significant differences with #p < 0.001 are indicated by hash key (#). Differences of means to control were calculated for all bars and all were significantly different from control with *p < 0.05 indicated by the lines on top of the bars except those bars that are marked with ns = not significant (p > 0.05).
Table 2.
IC50 values for the new compounds A13 and H12, and for staurosporine.
Fig 4.
Multiparametric assay in OS cell lines and control cell types after compound A13 treatment.
A) Immunofluorescence images show pseudo-colored nuclei (blue), mitochondria (red) and lysed cells (yellow). Cell type is indicated on the left, control (DMSO) and compound concentrations are indicated on the top of the images. Scale bars are 50 μM. Graphs in B)-E) show results from automated image quantification. The parameters cell count (B), mitochondrial (mito) mass (C), nucleus area (μm2) (D), and membrane permeability (%) (E) were determined for treatment with compound A13 at 10 μM for 24h. Unpaired t test showed *p < 0.05, **p < 0.01, ***p < 0.001, and not significant (ns) for U2OS vs. hOB and HOS vs. hOB, respectively, and in E) #p < 0.001 relative to DMSO-treated control of the respective cell type (A13-). Data are expressed as means +/- SD from duplicates of two independent experiments and are shown as percent or as fold change relative to DMSO-treated control.
Fig 5.
Multiparametric assay in OS cell lines and control cell types after compound H12 treatment.
A) Immunofluorescence images show pseudo-colored nuclei (blue), mitochondria (red) and lysed cells (yellow). Cell type is indicated on the left, control (DMSO) and compound concentrations on the top of the images. Scale bars are 50 μM. B)-E) shows results from automated image quantification. The parameters cell count (B), mitochondrial (mito) mass (C), nucleus area (μm2) (D), and membrane permeability (%) (E) were analyzed after treatment with compound H12 at 10 μM for 24h. Unpaired t test showed *p < 0.05, **p < 0.01, ***p < 0.001, and not significant (ns) for U2OS vs. hOB and HOS vs. hOB, respectively, and in E) #p < 0.001 relative to DMSO-treated control of the respective cell type (H12-). Data are expressed as means +/- SD from duplicates of two independent experiments and are shown as percent or as fold change relative to DMSO-treated control.
Fig 6.
Induction of caspase 3 and 7 activity and cell lysis in U2OS cell line after compound treatment.
A) shows caspase 3/7 activity and B) cell lysis for U2OS cell line treated with increasing concentrations of compound A13, H12, staurosporine and doxorubicin, respectively. Caspase 3/7 activity was determined after 8 and 24h of treatment using Caspase-Glo 3/7 Assay. Cell lysis was determined using CellTox Green Cytotoxicity Assay. Data are shown as fold change of mean +/- SD from duplicates of two independent experiments relative to DMSO-treated control. Unpaired t test was performed versus DMSO-treated control with *p < 0.05, **p < 0.01, and ***p < 0.001. C) shows cell count determined by quantification of Hoechst-stained nuclei after treatment of HOS cells with 20 μM of compound A13 and H12, respectively, and after co-treatment of cells with compounds and caspase inhibitors (pan-caspase inhibitor Z-VAD-FMK, caspase 8 inhibitor Z-IETD-FMK or caspase 9 inhibitor Z-LEHD-FMK) for 24h. Means and SD from duplicates of one representative experiment of two are shown (*p < 0.05 versus DMSO treated control; #p < 0.05 versus cells treated with 20 μM of compound).