The Major Heat Shock Proteins, Hsp70 and Hsp90, in 2-Methoxyestradiol-Mediated Osteosarcoma Cell Death Model

Abstract

2-Methoxyestradiol is one of the natural 17β-estradiol derivatives and a potential novel anticancer agent currently being under evaluation in advanced phases of clinical trials. However, the mechanism of anticancer action of 2-methoxyestradiol has not been yet fully established. In our previous studies we have demonstrated that 2-methoxyestradiol selectively induces the expression and nuclear translocation of neuronal nitric oxide synthase in osteosarcoma 143B cells. Heat shock proteins (Hsps) are factors involved in the regulation of expression and activity of nitric oxide synthases. Herein, we chose osteosarcoma cell lines differed in metastatic potential, metastatic 143B and highly metastatic MG63.2 cells, in order to further investigate the anticancer mechanism of 2-methoxyestradiol. The current study aimed to determine the role of major heat shock proteins, Hsp90 and Hsp70 in 2-methoxyestradiol-induced osteosarcoma cell death. We focused on the implication of Hsp90 and Hsp70 in control under expression of neuronal nitric oxide synthase, localization of the enzyme, and further generation of nitro-oxidative stress. To give the insight into the role of Hsp90 in regulation of anticancer efficacy of 2-methoxyestradiol, we used geldanamycin as a potent Hsp90 inhibitor. Herein, we evidenced that inhibition of Hsp90 controls the protein expression of 2-methoxyestradiol-induced neuronal nitric oxide synthase and inhibits enzyme nuclear translocation. We propose that decreased level of neuronal nitric oxide synthase protein after a combined treatment with 2-methoxyestradiol and geldanamycin is directly associated with the accompanying upregulation of Hsp70 and downregulation of Hsp90. This interaction resulted in abrogation of anticancer efficacy of 2-methoxyestradiol by geldanamycin.

Keywords: 2-methoxyestradiol; geldanamycin; neuronal nitric oxide synthase; osteosarcoma.

Figure 1

Anti-proliferative effects of 2-methoxyestradiol (2-ME) and GA in 143B and MG63.2 cell lines. Anti-proliferative effect of 2-ME or GA, or a combination of both compounds in OS 143B and MG63.2 cells. OS 143B (A) and M63.2 (B) were treated with series of dilutions (0.8 μM–50 μM) of either 2-ME or GA, or a combination of both (molar ratio 1:1). The cell proliferation was consequently determined by MTT assay. Data are presented as the mean ± SE values form at least three independent experiments. The absence of error bar denotes a line thickness greater than the error. Data were analyzed by GraphPad Prism Software version 6.02 performing One-way ANOVA combined with Dunett’s Multiple Comparison Test. * p < 0.01 versus control.

Figure 2

Antagonistic effect between 2-ME and GA. Anti-proliferative potential of 2-ME (red line), GA (green line), and the combination of both compounds (blue line) was determined by MTT assay as described above. Consequently, median-effect plot (A), dose–effect curve (B), and Fa-CI plot (C) were evaluated by CalcuSyn software. Values are the mean ± SE from three independent experiments.

Figure 3

GA abrogates stimulatory effect of 2-ME on nNOS expression and nuclear translocation in 143B and MG63.2 cell lines. Total pool of nNOS in OS 143B (A), MG63.2 (B) cells and in nuclear extract of 143B cells (C) treated with either 2-ME or GA, or a combination of both compounds. 143B and MG63.2 OS cells were treated with vehiculum (Control), 2-ME (1 μM, 10 μM) or GA (2 μM, 4 μM) or combination of both for 8 h. The cells were then harvested and total level of nNOS was established by western blotting. The cytoplasmic and nuclear fractions were separated using the Nuclear Extract Kit (Active Motif, France) according to manufacturer’s protocol. The representative data from three experiments are presented.

Figure 4

GA abrogates stimulatory effect of 2-ME on nNOS expression and nuclear translocation in 143B and MG63.2 cell lines. nNOS protein level in OS 143B and MG63.2 cells treated with vehiculum (Control) or either 2-ME or GA, or a combination of both compounds. OS 143B cells (A) and MG63.2 (B) were treated for with either 2-ME (1 μM or 10 μM) or GA (2 μM or 4 μM), or a combination of both for 8 h. The nNOS protein level was then determined by immunofluorescence. Cell nuclei were shown in blue, nNOS immunoreactivity in green, merged images of both in light blue. Original magnification ×40. Each experiment was performed at least three times. The representative data are shown.

Figure 5

The total pool of nitro-oxidative stress in 143B (A–E) and MG63.2 cells (F–J) treated with either with 2-ME or GA, or a combination of both compounds. OS 143B cells (A–D,I) and MG63.2 (E–H,J) were treated for with vehiculum (Control) or either 2-ME (1 μM or 10 μM) or GA (2 μM or 4 μM), or a combination of both for 8 h. (I,J) present the data additionally using the pre-treatment with specific nNOS inhibitor (10 μM 4-AAPNT) in OS 143B (I) and MG63.2 (J) cells. The total pool of nitro-oxidative stress was evaluated using 2′,7′–dichlorofluorescin diacetate (DCF-DA) staining by means of flow cytometry. Data are presented as the mean ± SE values form at least three independent experiments. The absence of error bar denotes a line thickness greater than the error. Data were analyzed by GraphPad Prism Software version 6.02 performing one way ANOVA analyses followed by Tukey’s multiple comparison test. * p < 0.01, ** p < 0.001 versus control.

Figure 6

2-ME and GA control the gene expression of Hsp90 alpha and Hsp90 beta in OS 143B cells. OS 143B cells were treated with either 2-ME (1 μM or 10 μM) or GA (2 μM or 4 μM) or in a combination of both for 8 h. Next, Hsp90 alpha (A–D) and Hsp90 beta (E–H) gene expression was determined by means of Real Time PCR. Data are presented as the mean ± SE values form at least three independent experiments. The absence of error bar denotes a line thickness greater than the error. Data were analyzed by GraphPad Prism Software version 6.02 performing one way ANOVA analyses followed by Tukey’s multiple comparison test. * p < 0.01, ** p < 0.001, *** p < 0.0001 versus control.

Figure 7

2-ME and GA regulates the Hsp90 protein level in OS 143B cells. OS 143B cells were treated with vehiculum (Control), 2-ME (1 μM, 10 μM) or GA (2 μM, 4 μM) or combination of both for 8 h. Consequently, total pool of Hsp90 protein in OS 143B was determined by western blotting (A) and immunofluorescence (B), respectively. Cell nuclei were shown in blue, Hsp90 immunoreactivity in red, merged images of both in light red. Original magnification ×40. While the protein level of Hsp90 alpha (C) and Hsp90 beta (D) was evaluated by western blotting. Each experiment was performed at least three times. The representative data from three experiments are presented.

Figure 8

2-ME has no effect on Hsp70, while GA controls the gene and protein expression of Hsp70 in OS 143B cells. OS 143B cells were treated with either 2-ME (1 μM or 10 μM) or GA (2 μM or 4 μM) or in a combination of both for 8 h. Next, Hsp70 (A–D) gene expression was determined by means of Real Time PCR. Consequently, total pool of Hsp70 protein in OS 143B was determined by western blotting (E) and immunofluorescence (F), respectively. Cell nuclei were shown in blue, Hsp70 immunoreactivity in green, merged images of both in light green. Original magnification ×40. The representative data from three experiments are presented. Data are presented as the mean ± SE values form at least three independent experiments. The absence of error bar denotes a line thickness greater than the error. Data were analyzed by GraphPad Prism Software version 6.02 performing one way ANOVA analyses followed by Tukey’s multiple comparison test. ** p < 0.001, *** p < 0.0001, **** p < 0.00001 versus control.