*2.5. AKG Inhibits the Migration and Invasiveness of OS Cells and Decreases the Production of VEGF and TGF-*β *in These Cells*

Since OS is classified as a strong tumor metastatic disease, the effect of AKG (5, 10, 25, 50 mM) on the migration (evaluated in a wound-healing assay) and invasion (evaluated in a transwell chamber assay with a basement membrane extract (BME)-coated membrane) of the Saos-2 and HOS cells was assessed. As shown in Figure 7A–D, the AKG treatment suppressed cell migration in both cell lines in a concentration-dependent manner. The inhibition of the migration of Saos-2 and HOS cells after the 24-h AKG treatment at a concentration of 50 mM increased approx. 2.3 and 2.5 times, respectively, in comparison with the migratory potential of control cells. At the same time, the AKG treatment decreased the invasiveness of both cell lines in a concentration-dependent manner (Figure 8A,B). The lowest AKG concentration used, i.e., 5 mM decreased the invasive activity of the Saos-2 and HOS cells by 14 ± 0.87% and 17 ± 2.71%, respectively, in comparison with the control cells. In turn, at the highest AKG concentration, i.e., 50 mM, the invasiveness of the Saos-2 and HOS cells decreased markedly by 43 ± 1.37% and 60.5 ± 3.53%, respectively. *Int. J. Mol. Sci.* **2020**, *21*, x FOR PEER REVIEW 7 of 21

**Figure 4.** Flow cytometry analysis of active caspase-3 in Saos-2 and HOS cells treated with AKG for 72 h. Representative histograms of Saos-2 (**A**) and HOS (**C**) cell cultures. Symbols M1 and M2 represent peaks for viable (caspase-3 negative cells) and apoptotic cell fractions (caspase-3 positive cells), respectively. Quantification of caspase-3 activity in Saos-2 (**B**) and HOS (**D**) cell cultures. Mean ± SD of 3 measurements in three independent experiments (*n* = 12 for each concentration); statistically **Figure 4.** Flow cytometry analysis of active caspase-3 in Saos-2 and HOS cells treated with AKG for 72 h. Representative histograms of Saos-2 (**A**) and HOS (**C**) cell cultures. Symbols M1 and M2 represent peaks for viable (caspase-3 negative cells) and apoptotic cell fractions (caspase-3 positive cells), respectively. Quantification of caspase-3 activity in Saos-2 (**B**) and HOS (**D**) cell cultures. Mean ± SD of 3 measurements in three independent experiments (*n* = 12 for each concentration); statistically significant at *p* < 0.001 \*\*\* in comparison to the control; one-way ANOVA test.

significant at *p* < 0.001 \*\*\* in comparison to the control; one-way ANOVA test.

of pro-apoptotic signals through an increase in the Bax/Bcl-2 ratio.

A further study was undertaken to resolve which pathway, receptor- or mitochondriadependent, was involved in the AKG-induced caspase 3 activation in the OS cells. Since a higher number of cells with activated caspase 3 was observed in the Saos-2 culture, this cell line was chosen to examine the activation of initiator caspase-8 (extrinsic pathway) and caspase-9 (intrinsic pathway) with the use of immunoblotting and flow cytometry methods. As shown in Figure 5A–C, AKG caused a slight increase in the active forms of caspase-8, in comparison with the control, after the 72-h treatment, although only in a small percentage of cells. In contrast, an AKG concentration-dependent decrease in the procaspase-9 levels and an increase in its active form were observed in the Saos-2 cells (Figure 5D), and a large number of cells exhibited the presence of the active form of this caspase In addition, the production of some cell migration-, invasion- and angiogenesis-associated growth factors such as transforming growth factor β (TGF-β) and vascular endothelial growth factor (VEGF) was examined by ELISA. As shown in Figure 8C,D, both OS cell lines constitutively produced significant amounts of TGF-β, although the Saos-2 cells secreted almost two-fold higher levels of this cytokine than the HOS cells. The level of TGF-β produced by the control Saos-2 and HOS cultures was 5570 ± 27.85 pg/mL and 2851 ± 55.70 pg/mL, respectively. The 72-h treatment of the Saos-2 and HOS cells with AKG suppressed the production of TGF-β in a concentration-dependent manner in both OS cell types (Figure 8C,D). At the highest concentration tested, i.e., 50 mM, AKG decreased the production of TGF-β in the Saos-2 and HOS cells by approx. 29% and 43%, respectively. Similarly, both OS cell lines secreted constitutively significant amounts of VEGF, although also in this case the

inhibitor) in the Saos-2 cells. This suggests that the AKG treatment may result in the predominance

(Figure 5E,F). Since these results indicated that the AKG-treatment activated predominantly the intrinsic apoptotic pathway, the expression of pro-apoptotic and anti-apoptotic proteins associated control Saos-2 cells produced substantially higher quantities of this growth factor than the HOS cells (29 670 ± 35 pg/mL vs. 1985 ± 14 pg/mL, respectively). The 72-h treatment of the Saos-2 and HOS cells with AKG suppressed the production of VEGF in a concentration-dependent manner in both OS cell lines (Figure 8E,F); however, this effect was stronger in the HOS cells. In this cell line, all the AKG concentrations tested, i.e., 5, 10, 25, and 50 mM, inhibited significantly VEGF production by approx. 21%, 33%, 74%, and 94%, respectively (Figure 8F). After the treatment of the Saos-2 cells with AKG at a concentration of 10, 25, and 50 mM, the production of VEGF decreased by approx. 9%, 17%, and 39%, respectively (Figure 8E). *Int. J. Mol. Sci.* **2020**, *21*, x FOR PEER REVIEW 8 of 21

**Figure 5.** Analysis of the expression of apoptosis-related proteins in Saos-2 cells treated with AKG. After 72-h incubation with AKG, the expressions of procaspases-8 and -9 and cleaved forms of these caspases were examined by Western blotting, and active caspase-8 and -9 were analyzed by flow cytometric analysis. Representative blots from three independent experiments (**A**,**D**). Representative histograms of Saos-2 cell culture (**B**,**E**). Symbol M1 represents peaks for active caspase-8 or -9 positive cells. Quantification of caspase-8 (**C**) and -9 (**F**) activity in Saos-2 cell cultures. Mean ± SD of 3 measurements in three independent experiments (*n* = 12 for each concentration); statistically significant at \*\*\* *p* < 0.001 in comparison to the control; one-way ANOVA test. Western blotting of Bax and Bcl-2 expression after 72-h treatment with AKG (**G**) Equal loading was confirmed by immunodetection of *β*-actin. Densitometry analysis of Bax and Bcl-2 bands with ImageLab™ Software (**H**). Data are expressed as means ± SD for at least three independent experiments; (*n* = 3), \* *p* < 0.05 **Figure 5.** Analysis of the expression of apoptosis-related proteins in Saos-2 cells treated with AKG. After 72-h incubation with AKG, the expressions of procaspases-8 and -9 and cleaved forms of these caspases were examined by Western blotting, and active caspase-8 and -9 were analyzed by flow cytometric analysis. Representative blots from three independent experiments (**A**,**D**). Representative histograms of Saos-2 cell culture (**B**,**E**). Symbol M1 represents peaks for active caspase-8 or -9 positive cells. Quantification of caspase-8 (**C**) and -9 (**F**) activity in Saos-2 cell cultures. Mean ± SD of 3 measurements in three independent experiments (*n* = 12 for each concentration); statistically significant at \*\*\* *p* < 0.001 in comparison to the control; one-way ANOVA test. Western blotting of Bax and Bcl-2 expression after 72-h treatment with AKG (**G**) Equal loading was confirmed by immunodetection of β-actin. Densitometry analysis of Bax and Bcl-2 bands with ImageLab™ Software (**H**). Data are expressed as means ± SD for at least three independent experiments; (*n* = 3), \* *p* < 0.05 and \*\*\* *p* < 0.001 in comparison to the control; one-way ANOVA test.

*2.4. AKG Modulates the Phosphorylation of Mitogen-Activated Protein Kinases and Induces Apoptosis in* 

To explore the involvement of mitogen-activated protein kinases (MAPKs) in AKG-induced OS cell apoptosis, phosphorylation of JNK, extracellular signal-regulated kinase (ERK1/2), and p38 was examined with the quantitative ELISA method. As shown in Figure 6A, the AKG treatment reduced ERK1/2 phosphorylation in a concentration-dependent manner within 6 and 24 h in the Saos-2 cells. In contrast, AKG remarkably augmented the level of phospho-JNK in a concentration-dependent

JNK is a stress-activated kinase, and a signaling pathway with the participation of this kinase

and \*\*\* *p* < 0.001 in comparison to the control; one-way ANOVA test.

manner, but not phospho-p38 (Figure 6B,C).

*OS Cells through a c-Jun N-terminal protein kinase (JNK)-Dependent Mechanism* 

inhibitor of JNK (SP600125). After 72 h, the percentage of apoptotic cells was measured with FACS, whereas the JNK phosphorylation status was evaluated after 24 h with ELISA. As shown in Figure 6D, the percentage of apoptotic cells declined from 22.52 ± 1.8% after the treatment with 50 mM AKG alone to 13.1 ± 1.2% when the cells were co-treated with AKG and 5 μM of SP600125. It was found that the inhibition of JNK phosphorylation by SP600125 completely inhibited the activation of JNK induced by this compound (Figure 6E) and partially reduced the level of AKG-induced apoptosis in

*TGF-β in These Cells* 

pathway.

the Saos-2 cells (Figure 6F). These data may therefore support the observation that the AKG-induced

**Figure 6***.* Effect of AKG on phosphorylation of MAP kinases and the influence of the selective JNK inhibitor (SP600125) on AKG-induced apoptosis in Saos-2 cells. The cells were incubated without or with AKG for 6 h and 24 h, and phosphorylated and total ERK1/2, JNK and p38 levels were determined with the ELISA assay. Quantification of the amounts of phosphorylated to total MAP kinases (**A**–**C**). The cells were treated with 50 mM AKG without or with SP600125 (5 μM) and harvested after 72 h of treatment for apoptosis analysis. The representative dot plots indicate the percentage of An<sup>−</sup>/PI+ necrotic cells (Q1), An+/PI+ late apoptotic cells (Q2), An<sup>−</sup>/PI<sup>−</sup> viable cells (Q3), and An+/PI<sup>−</sup> early apoptotic cells (Q4) in the AKG or/and SP600125-treated Saos-2 cell cultures (**D**). Quantification of the amounts of phosphorylated to total JKN kinase (**E**) and histogram representation of the quantitative percentage of apoptotic (early + late apoptosis) cells (**F**) in the control, SP600125, AKG, and SP600125 + AKG-treated Saos-2 cell cultures. Data are expressed as means ± SD for three independent experiments. \*\* *p* < 0.01, \*\*\* *p* < 0.001 in comparison to the control; one-way ANOVA test. **Figure 6.** Effect of AKG on phosphorylation of MAP kinases and the influence of the selective JNK inhibitor (SP600125) on AKG-induced apoptosis in Saos-2 cells. The cells were incubated without or with AKG for 6 h and 24 h, and phosphorylated and total ERK1/2, JNK and p38 levels were determined with the ELISA assay. Quantification of the amounts of phosphorylated to total MAP kinases (**A**–**C**). The cells were treated with 50 mM AKG without or with SP600125 (5 µM) and harvested after 72 h of treatment for apoptosis analysis. The representative dot plots indicate the percentage of An−/PI<sup>+</sup> necrotic cells (Q1), An+/PI<sup>+</sup> late apoptotic cells (Q2), An−/PI<sup>−</sup> viable cells (Q3), and An+/PI<sup>−</sup> early apoptotic cells (Q4) in the AKG or/and SP600125-treated Saos-2 cell cultures (**D**). Quantification of the amounts of phosphorylated to total JKN kinase (**E**) and histogram representation of the quantitative percentage of apoptotic (early + late apoptosis) cells (**F**) in the control, SP600125, AKG, and SP600125 + AKG-treated Saos-2 cell cultures. Data are expressed as means ± SD for three independent experiments. \*\* *p* < 0.01, \*\*\* *p* < 0.001 in comparison to the control; one-way ANOVA test.

*2.5. AKG Inhibits the Migration and Invasiveness of OS Cells and Decreases the Production of VEGF and* 

Since OS is classified as a strong tumor metastatic disease, the effect of AKG (5, 10, 25, 50 mM) on the migration (evaluated in a wound-healing assay) and invasion (evaluated in a transwell chamber assay with a basement membrane extract (BME)-coated membrane) of the Saos-2 and HOS cells was assessed. As shown in Figure 7A–D, the AKG treatment suppressed cell migration in both cell lines in a concentration-dependent manner. The inhibition of the migration of Saos-2 and HOS cells after the 24-h AKG treatment at a concentration of 50 mM increased approx. 2.3 and 2.5 times, respectively, in comparison with the migratory potential of control cells. At the same time, the AKG treatment decreased the invasiveness of both cell lines in a concentration-dependent manner (Figure

control; one-way ANOVA test.

8A,B). The lowest AKG concentration used, i.e., 5 mM decreased the invasive activity of the Saos-2 and HOS cells by 14 ± 0.87% and 17 ± 2.71%, respectively, in comparison with the control cells. In

decreased markedly by 43 ± 1.37% and 60.5 ± 3.53%, respectively.

**Figure 7.** Effect of AKG on migration of Saos-2 and HOS cells in the wound healing assay. Cells were scraped and cultured without or with AKG for 24 h. Saos-2 (**A**) and HOS (**C**) cultures were imaged under a contrast-phase light microscope (magnification ×40) before and after injury. Cell migration was quantified by measuring the gap width of wounds. Quantitative data (**B**,**D**) are presented as a relative fold change in the inhibition of migration in comparison to the control. Data are expressed as means ± SD for three independent experiments. scale bar = 200 μm,\*\*\* *p* < 0.001 in comparison to the **Figure 7.** Effect of AKG on migration of Saos-2 and HOS cells in the wound healing assay. Cells were scraped and cultured without or with AKG for 24 h. Saos-2 (**A**) and HOS (**C**) cultures were imaged under a contrast-phase light microscope (magnification ×40) before and after injury. Cell migration was quantified by measuring the gap width of wounds. Quantitative data (**B**,**D**) are presented as a relative fold change in the inhibition of migration in comparison to the control. Data are expressed as means ± SD for three independent experiments. scale bar = 200 µm,\*\*\* *p* < 0.001 in comparison to the control; one-way ANOVA test.

In addition, the production of some cell migration-, invasion- and angiogenesis-associated growth factors such as transforming growth factor β (TGF-β) and vascular endothelial growth factor (VEGF) was examined by ELISA. As shown in Figure 8C,D, both OS cell lines constitutively produced significant amounts of TGF-β, although the Saos-2 cells secreted almost two-fold higher levels of this cytokine than the HOS cells. The level of TGF-β produced by the control Saos-2 and HOS cultures was 5570 ± 27.85 pg/mL and 2851 ± 55.70 pg/mL, respectively. The 72-h treatment of the Saos-2 and HOS cells with AKG suppressed the production of TGF-β in a concentration-dependent manner in both OS cell types (Figure 8C,D). At the highest concentration tested, i.e.*,* 50 mM, AKG decreased the production of TGF-β in the Saos-2 and HOS cells by approx. 29% and 43%, respectively. Similarly, both OS cell lines secreted constitutively significant amounts of VEGF, although also in this case the control Saos-2 cells produced substantially higher quantities of this growth factor than the HOS cells

**Figure 8.** Effect of AKG on cell invasion and production of transforming growth factor β (TGF-β) and vascular endothelial growth factor (VEGF) in Saos-2 and HOS cells. Cell invasion was evaluated in a transwell chamber assay with a basement membrane extract (BME)-coated membrane (8 µM) after 24 h. Invaded Saos-2 (**A**) and HOS (**B**) cells were quantified by measuring calcein-AM fluorescence. Following the 72-h AKG treatment, the conditioned media from the Saos-2 and HOS cell cultures were collected and the levels of TGF-β (**C**,**D**) and VEGF (**E**,**F**) were assayed with ELISA. Representative results of three independent experiments are shown. (*n* = 9); statistically significant at *p* < 0.05 \*, at *p* < 0.01 \*\* or at *p* < 0.001 \*\*\* in comparison to the control; one-way ANOVA test.
