Osteosarcoma Cells and Undifferentiated Human Mesenchymal Stromal Cells Are More Susceptible to Ferroptosis than Differentiated Human Mesenchymal Stromal Cells

Abstract

Current research suggests that promoting ferroptosis, a non-apoptotic form of cell death, may be an effective therapy for osteosarcoma, while its inhibition could facilitate bone regeneration and prevent osteoporosis. Our objective was to investigate whether the susceptibility to and regulation of ferroptosis differ between undifferentiated (UBC) and differentiated (DBC) human bone marrow stromal cells, as well as human osteosarcoma cells (MG63). Ferroptosis was induced by either inhibiting glutathione peroxidase 4 (GPX4) using RSL3 or blocking all glutathione-dependent enzymes through inhibition of the glutamate/cysteine antiporter with Erastin. Lipid peroxidation was assessed using the fluorescent probe BODIPY™581/591C11, while Ferrostatin-1 was used to inhibit ferroptosis. We demonstrate that neither Erastin nor RSL3 induces ferroptosis in DBC. However, both RSL3 and Erastin induce ferroptosis in UBC, while Erastin predominantly induces ferroptosis in MG63 cells. Our data suggest that ferroptosis induction in undifferentiated hBMSCs is primarily regulated by GPX4, whereas glutathione S-Transferase P1 (GSTP1) plays a key role in controlling ferroptosis in osteosarcoma cells. In conclusion, targeting the key pathways involved in ferroptosis across different bone cell types may improve the efficacy of cancer treatments while minimizing collateral damage and supporting regenerative processes, with minimal impact on cancer therapy.

Keywords: Erastin; Ferrostatin-1; RSL3; ferroptosis; osteosarcoma.

Figure 1

An example illustrating the calculation of the area under the dose curve (AUDC) in representative images. (A) AUDC is the integral area calculated under the curve of either LDH release or values LPO at all concentrations of ferroptosis inducers. Thus, this single parameter comprises the results obtained with all concentrations of either Erastin and or RSL3 representing an integrative parameter characterizing efficiency of both substances to induce ferroptosis. Such a parameter is particularly required if the cells strongly react to the small changes in inducer concentrations as it is the case in our study. (B) Determination of the levels of lipid peroxidation. The level of lipid peroxidation (% oxidized lipids) was determined from images as shown in B based on the determination of the fluorescens of oxidized (green) and non-oxidized (red) lipids in the BODIPY™ 581/591 C11 stained cells. The % oxidized lipids was calculated using the following equation: % oxidized lipids = intensity green/(intensity red + intensity green). Fluorescence analysis shows 1 representative set of images out of 4 independent experiments. Scale bar: 140 μm.

Figure 2

Baseline LDH release (A) and LPO levels (B) in three types of untreated cells. The cells were incubated in fresh incubation medium for 24 h. After 24 h the medium was collected to determine the LDH levels and cells were stained with BODIPY™ 581/591 C11 to determine lipid peroxidation. *—p ≤ 0.05. Statistical evaluation was performed with one way ANOVA followed by post hoc Holm–Sidak’s multiple comparisons test.

Figure 3

Effect of Erastin on undifferentiated human bone marrow mesenchymal stem cells with and without the addition of Ferrostatin-1. (A) Effect of Erastin on changes in cell morphology. The images were taken with a Zeiss LSM 510 microscope, 10× lens (scale bar: 140 μm). (B) Changes in levels of LDH release under various conditions. (C) Effect of Erastin on lipid peroxidation staining with BODIPY™ 581/591 C11. The images were taken with a Zeiss LSM 510 microscope, 10× lens using a red color filter for the non-oxidized form of the dye and green color filter for the oxidized form of the dye, after that two images were merged (scale bar: 140 μm). (D) Effect of Erastin on lipid peroxidation. Data are represented as means ± SEM (error bars), statistical significance was analyzed with RM-ANOVA followed by post hoc Holm–Sidak’s multiple comparisons test. n = 4. *—p ≤ 0.05; **—p ≤ 0.001; ***—p ≤ 0.0001.

Figure 4

Effect of RSL3 on undifferentiated human bone marrow mesenchymal stem cells with and without the addition of Ferrostatin-1. (A) Effect of RSL3 on changes in cell morphology. The images were taken with a Zeiss LSM 510 microscope, 10× lens (scale bar: 140 μm). (B) Changes in levels of LDH release under various conditions. (C) Effect of RSL3 on lipid peroxidation staining with BODIPY™ 581/591 C11. The images were taken with a Zeiss LSM 510 microscope, 10× lens using a red color filter for the non-oxidized form of the dye and green color filter for the oxidized form of the dye, after the two images were merged (scale bar: 140 μm). (D) Effect of RSL3 on lipid peroxidation. Data are represented as means ± SEM (error bars), statistical significance was analyzed with RM-ANOVA followed by post hoc Holm–Sidak’s multiple comparisons test. n = 4. *—p ≤ 0.05; **—p ≤ 0.01, ***—p ≤ 0.001, ****—p ≤ 0.0001.

Figure 5

Effect of Erastin on differentiated human bone marrow mesenchymal stem cells (DBC) with and without the addition of Ferrostatin-1. (A) Effect of Erastin on changes in cell morphology. The images were taken with a Zeiss LSM 510 microscope, 10× lens (scale bar: 140 μm). (B) Changes in levels of LDH release under various conditions. (C) Effect of Erastin on lipid peroxidation staining with BODIPY™ 581/591 C11. The images were taken with a Zeiss LSM 510 microscope, 10× lens using a red color filter for the non-oxidized form of the dye and green color filter for the oxidized form of the dye, after that two images were merged (scale bar: 140 μm). (D) Effect of Erastin on lipid peroxidation. Data are represented as means ± SEM (error bars), statistical significance was analyzed with RM-ANOVA followed by post hoc Holm–Sidak’s multiple comparisons test. n = 6.

Figure 6

Effect of RSL3 on differentiated human bone marrow stromal cells (DBC) with and without the addition of Ferrostatin-1. (A) Effect of RSL3 on changes in cell morphology. The images were taken with a Zeiss LSM 510 microscope, 10× lens (scale bar: 140 μm). (B) Changes in levels of LDH release under various conditions. (C) Effect of RSL3 on lipid peroxidation staining with BODIPY™ 581/591 C11. The images were taken with a Zeiss LSM 510 microscope, 10× lens using a red color filter for the non-oxidized form of the dye and green color filter for the oxidized form of the dye, after that two images were merged (scale bar: 140 μm). (D) Effect of RSL3 on lipid peroxidation. Data are represented as means ± SEM (error bars), statistical significance was analyzed with RM-ANOVA followed by post hoc Holm–Sidak’s multiple comparisons test. n = 6. *—p ≤ 0.05.

Figure 7

Effect of Erastin on human osteosarcoma cell line MG63 with and without the addition of Ferrostatin-1. (A) Effect of Erastin on changes in cell morphology. The images were taken with a Zeiss LSM 510 microscope, 10× lens (scale bar: 140 μm). (B) Changes in levels of LDH release under various conditions. (C) Effect of Erastin on lipid peroxidation staining with BODIPY™ 581/591 C11. The images were taken with a Zeiss LSM 510 microscope, 10× lens using a red color filter for the non-oxidized form of the dye and green color filter for the oxidized form of the dye, after that two images were merged (scale bar: 140 μm). (D) Effect of Erastin on lipid peroxidation. Data are represented as means ± SEM (error bars), statistical significance was analyzed with RM-ANOVA followed by post hoc Holm–Sidak’s multiple comparisons test. n = 4. *—p ≤ 0.05; **—p ≤ 0.01, ***—p ≤ 0.001, ****—p ≤ 0.0001.

Figure 8

Effect of RSL3 on human osteosarcoma cell line MG63 with and without the addition of Ferrostatin-1. (A) Effect of RSL3 on changes in cell morphology. The images were taken with a Zeiss LSM 510 microscope, 10× lens (scale bar: 140 μm). (B) Changes in levels of LDH release under various conditions. (C) Effect of RSL3 on lipid peroxidation staining with BODIPY™ 581/591 C11. The images were taken with a Zeiss LSM 510 microscope, 10× lens using a red color filter for the non-oxidized form of the dye and green color filter for the oxidized form of the dye, after that two images were merged (scale bar: 140 μm). (D) Effect of RSL3 on lipid peroxidation. Data are represented as means ± SEM (error bars), statistical significance was analyzed with RM-ANOVA followed by post hoc Holm–Sidak’s multiple comparisons test. n = 4. *—p ≤ 0.05, ****—p ≤ 0.0001.

Figure 9

Area under dose-response curve for LDH analysis. (A) Effect of Erastin on UBC. (B) Effect of Erastin on DBC. (C) Effect of Erastin on MG63. (D) Effect of RSL3 on UBC. (E) Effect of RSL3 on DBC. (F) Effect of RSL3 on MG63. The data are presented as means ± SEM (error bars). Statistical evaluation was performed with one way ANOVA followed by post hoc Holm–Sidak’s multiple comparisons test. *—p ≤ 0.05; **—p ≤ 0.01.

Figure 10

Area under dose-response curve for LPO analysis. (A) Effect of Erastin on UBC. (B) Effect of Erastin on DBC. (C) Effect of Erastin on MG63. (D) Effect of RSL3 on UBC. (E) Effect of RSL3 on DBC. (F) Effect of RSL3 on MG63 Data are presented as means ± SEM (error bars). Statistical evaluation was performed with one way ANOVA followed by post hoc Holm–Sidak’s multiple comparisons test. *—p ≤ 0.05; **—p ≤ 0.01; ***—p ≤ 0.001.

Figure 11

Mode of action of RSL3 and Erastin in tested bone cells. RSL3 inhibits specifically GPX4 and is expected to activate ferroptosis if the levels of LOOH are controlled predominantly by GPX4. Erastin inhibits glutamate-cysteine antiporter and is expected to activate ferroptosis if it is controlled either by GPX4 or by other GSH dependent enzymes, such as GSTP1. The fact that both RSL3 and Erastin induce ferroptosis in UBC suggests that it is controlled by GPX4. The fact that in MG63 ferroptosis is activated only by Erastin suggest that it is controlled by GSTP1. The elevated synthesis of LOOH is the obligatory prerequisite for induction of ferroptosis. If the initial levels of LOOH is low, as it was observed in DBC, then ferroptosis cannot be executed. This scheme illustrates the two major effector pathways.