Intracellular Signaling Responses Induced by Radiation within an In Vitro Bone Metastasis Model after Pre-Treatment with an Estrone Analogue

2-Ethyl-3-O-sulfamoyl-estra-1,3,5(10)16-tetraene (ESE-16) is an in silico-designed estradiol analogue which has improved the parent compound’s efficacy in anti-cancer studies. In this proof-of-concept study, the potential radiosensitizing effects of ESE-16 were investigated in an in vitro deconstructed bone metastasis model. Prostate (DU 145) and breast (MDA-MB-231) tumor cells, osteoblastic (MC3T3-E1) and osteoclastic (RAW 264.7) bone cells and human umbilical vein endothelial cells (HUVECs) were representative components of such a lesion. Cells were exposed to a low-dose ESE-16 for 24 hours prior to radiation at non-lethal doses to determine early signaling and molecular responses of this combination treatment. Tartrate-resistant acid phosphatase activity and actin ring formation were investigated in osteoclasts, while cell cycle progression, reactive oxygen species generation and angiogenic protein expression were investigated in HUVECs. Increased cytotoxicity was evident in tumor and endothelial cells while bone cells appeared to be spared. Increased mitotic indices were calculated, and evidence of increased deoxyribonucleic acid damage with retarded repair, together with reduced metastatic signaling was observed in tumor cells. RAW 264.7 macrophages retained their ability to differentiate into osteoclasts. Anti-angiogenic effects were observed in HUVECs, and expression of hypoxia-inducible factor 1-α was decreased. Through preferentially inducing tumor cell death and potentially inhibiting neovascularization whilst preserving bone physiology, this low-dose combination regimen warrants further investigation for its promising therapeutic application in bone metastases management, with the additional potential of limited treatment side effects.


Introduction
Metastasis is an inefficient process with an estimate of only 0.01% of tumor cells successfully disseminating from a solid tumor and establishing themselves at a distant site [1]. However, bone metastasis is highly prevalent in breast and prostate cancer patients [2]. Establishment of bone metastatic lesions involves a complex interaction between tumor cells, bone-forming and -resorbing cells, angiogenesis, and mineralized bone matrix to form a supportive microenvironment to foster secondary tumor formation [1,3,4].
Metastasis relies on the migratory and invasive properties of propagating tumor cells to form distant lesions. Undifferentiated tumor cells undergo epithelial-mesenchymal transition (EMT) which is brought about by genetic and epigenetic signaling in the tumor microenvironment [5]. Tumor-associated neovasculature is essential for the delivery of nutrients and oxygen to the tumor microenvironment and the removal of metabolic waste one of these sulfamoylated 2ME2 analogues that is not commercially available ( Figure 1) [38,39]. ESE-16 is not a substrate for P-glycoprotein pumps, acts independently of hormone receptors, and demonstrates unchanged oral absorption in murine models [33,40]. ESE-16 has demonstrated anti-tumorigenic effects at nanomolar concentrations in vitro through a crosstalk mechanism involving apoptosis and autophagy [39,41]. The estrone analogue disrupts microtubule dynamics and causes metaphase arrest in various cancer cell lines, including oestrogen receptor-positive MCF-7 and oestrogen receptornegative MDA-MB-231 breast cancer cells, whilst sparing non-tumorigenic cells as well as erythrocytes, fibrin networks, and platelets [38,[41][42][43]. Radiation is frequently used in the treatment of bone metastases. However, it is accompanied by general and skeletal adverse effects. Using MTAs as radiosensitizers may augment nuclear damage and delay the repair thereof by impairing nucleocytoplasmic shuttling of DNA damage and repair signaling proteins [44,45]. The effects of the ESE-16 in combination with radiation has not been investigated in tumorigenic prostate cells and non-tumorigenic cell lines such as those of the bone and endothelium. The aim of this study was to investigate the cellular responses of breast and prostate cancer bone metastases receiving radiation following pre-sensitization with ESE-16 using a deconstructed in vitro model comprised of tumor, bone, and endothelial cells. This proofof-concept study investigated the individual responses of the representative components of bone metastasis and provides insight into the independent intracellular signaling pathways as well as the potential modification of the ECM.

Cell Lines
Metastatic human prostate carcinoma cells (DU 145) and breast adenocarcinoma cells (MDA-MB-231); murine pre-osteoblasts (MC3T3-E1) and macrophages (RAW 264.7); and human umbilical vein endothelial cells (HUVECs) were used in this in vitro study. All cell lines are commercially available and were purchased from American Type Culture Collection (ATCC) (Maryland (MD), United States of America (USA)). Stander et al. designed a range of 2ME2 analogues in silico in an attempt to overcome the pharmacokinetic constraints of the parent compound by improving its bioavailability, potency, and tumor localization due to preferential carbonic anhydrase IX binding in the acidic micromilieu [33,36,37]. 2-Ethyl-3-O-sulfamoyl-estra-1,3,5(10)16-tetraene (ESE-16) is one of these sulfamoylated 2ME2 analogues that is not commercially available ( Figure 1) [38,39]. ESE-16 is not a substrate for P-glycoprotein pumps, acts independently of hormone receptors, and demonstrates unchanged oral absorption in murine models [33,40]. ESE-16 has demonstrated anti-tumorigenic effects at nanomolar concentrations in vitro through a crosstalk mechanism involving apoptosis and autophagy [39,41]. The estrone analogue disrupts microtubule dynamics and causes metaphase arrest in various cancer cell lines, including oestrogen receptor-positive MCF-7 and oestrogen receptor-negative MDA-MB-231 breast cancer cells, whilst sparing non-tumorigenic cells as well as erythrocytes, fibrin networks, and platelets [38,[41][42][43].
Radiation is frequently used in the treatment of bone metastases. However, it is accompanied by general and skeletal adverse effects. Using MTAs as radiosensitizers may augment nuclear damage and delay the repair thereof by impairing nucleocytoplasmic shuttling of DNA damage and repair signaling proteins [44,45]. The effects of the ESE-16 in combination with radiation has not been investigated in tumorigenic prostate cells and nontumorigenic cell lines such as those of the bone and endothelium. The aim of this study was to investigate the cellular responses of breast and prostate cancer bone metastases receiving radiation following pre-sensitization with ESE-16 using a deconstructed in vitro model comprised of tumor, bone, and endothelial cells. This proof-of-concept study investigated the individual responses of the representative components of bone metastasis and provides insight into the independent intracellular signaling pathways as well as the potential modification of the ECM.

Compound and Experimental Protocols
ESE-16 is not commercially available and was synthesized by iThemba Pharmaceuticals (Pty) Ltd. (Johannesburg, Gauteng, South Africa (SA)). ESE-16 was dissolved in dimethyl sulfoxide (DMSO) (Sigma-Aldrich, St Louis, MA, USA) to a 10-mM stock solution and stored at −20 • C. Exponentially growing cells were seeded at appropriate densities for each experiment, with a 24-h cell attachment policy. The lowest ESE-16 concentration to significantly induce apoptosis and the lowest radiation dose to significantly reduce cell viability were determined in the experimental dose-response curve set-up (Supporting data). DU 145 cells, RAW 264.7 macrophages, and HUVECs were exposed to 0.235 µM ESE-16 and MDA-MB-231 cells were exposed to 0.176 µM of ESE 16 for 24 h. After the 24-h ESE-16-exposure, cells were exposed to a single dose of 4 Gy X-radiation. Termination proceeded 2, 24, and 48 h after irradiation.

Radiation
Samples were irradiated using a Siemens Oncor Impression Linear Accelerator (Siemens Medical Solutions, Malvern, Pennsylvania (PA), USA) at the Department of Radiation Oncology, Steve Biko Academic Hospital (Pretoria, Gauteng, SA). Photon beam energy of 6 MV was used with a direct field and a gantry angle of 180 • . Source-to-surface distance (SSD) was 100 cm and the photon beam electronic equilibrium depth (Dmax) was 1.5 cm. A field size of 10 × 10 cm was used with 4.5-cm tissue equivalent bolus for dose homogeneity.

Controls
Cells propagated in DMEM served as a negative control and DMSO (0.05% v/v) served as a vehicle control. No statistical differences were observed between the negative and vehicle controls. The ESE-16 treatment control entailed ESE-16 exposure at 24 h after seeding for the experimental duration. The radiation treatment control followed the same experimental timeline with irradiation at 48 h after seeding. Paclitaxel served as a positive method control for apoptosis, 2ME2 for mitotic arrest, and etoposide for DNA damage.

Cell Division: Mitotic Index Determination Using Light Microscopy
Cells were fixed with Bouin's solution (Sigma-Aldrich, St Louis, MO, USA) (RT, 30 min), followed by washing with 70% ethanol (20 min) and double-distilled water (ddH 2 O). Cells were stained with Harris modified hematoxylin solution (Sigma-Aldrich, St Louis, MO, USA) (RT, 20 min), followed by washing with ddH 2 O and 70% ethanol. Cells were stained with 1% eosin (British Drug Houses, Ltd., London, UK) in 70% ethanol (RT, 7 min). Cells were washed twice (5 min each) with ethanol (70%, 96%, 100%) and xylol. Entellan ® mounting fluid (Merck, Darmstadt, Germany) was used to mount coverslips onto slides. A Nikon Optiphot transmitted light microscope (Tokyo, Japan) was used for visualization. The cells were scored blind, with slides deidentified, and were only linked to each treatment condition after the analysis was concluded.

Apoptosis: Annexin V-FITC Detection by Flow Cytometry
The Annexin V-fluorescein isothiocyanate (FITC) Apoptosis Detection Kit (BioLegend, Inc., San Diego, CA, USA) was used. Trypsinized cells were resuspended in 100 µL of Annexin V binding buffer, to which 5 µL of Annexin V-FITC and 10 µL of PI were added (RT, 15 min, protected from light). An additional 400 µL of Annexin V binding buffer was added. FITC (FL1) and PI (FL3) fluorescence were measured on a FC 500 series flow cytometer (Beckman Coulter, Brea, CA, USA).

Effect of the Combination Treatment on the Bone Component of the Metastatic Lesion
For TRAP activity and staining, RAW 264.7 macrophages were exposed to ESE-16 for 24 h prior to 4 Gy irradiation then differentiated over 5 days in the presence of 15 ng/µL RANKL. In parallel, RAW 264.7 macrophages were differentiated over 5 days in the presence of 15 ng/µL RANKL then treated with ESE-16 for 24 h followed by 4 Gy irradiation. For actin ring formation, RAW 264.7 macrophages were fully differentiated over 5 days in the presence of 30 ng/µL RANKL as previously described, then treated with ESE-16 for 24 h followed by 4 Gy irradiation [46]. Growth medium was replaced after 3 days at which another supplement of RANKL was added.

Statistical Analyses
Qualitative data were obtained from confocal and fluorescence microscopic techniques performed in triplicate. Semi-quantitative (mitotic index (MI), micronuclei (Mn) and TRAP staining), and quantitative data (spectrophotometry, flow cytometry, and Western blotting) were obtained from a minimum of three independent biological repeats (n ≥ 3) in the cytotoxicity studies. Flow cytometric data from a minimum of 10,000 cells was analyzed using Kaluza Analysis Software, version 2.0 (Beckman Coulter, Brea, CA, USA). Western blot analysis was performed on Image Lab, version 6.0 (Bio-Rad Laboratories, Inc., Hercules, CA, USA). Statistical analyses employed the analysis of variance (ANOVA)-single factor model and two-tailed Student's t-test. p-values < 0.05 were considered as statistically significant.

ESE-16 Is Cytotoxic to Cancer, Pre-Osteoclastic, and Endothelial Cells with Enhanced Sensitivity When Combined with Radiation Whilst Pre-Osteoblasts Are Spared
The GI 50 values of a 48-h ESE-16-exposure was determined spectrophotometrically with a dose-response curve in all cell lines (Table 1 and Figure 2). DU 145 and MDA-MB-231 cells yielded GI 50 values of 0.625 ± 0.039 µM and 0.469 ± 0.078 µM, respectively. RAW 264.7 macrophages yielded a GI 50 value of 0.625 ± 0.156 µM with increased sensitivity to 4 Gy radiation (GI 50 < 0.039 µM). HUVECS yielded a GI 50 value of 0.156 ± 0.02 µM with increased sensitivity to radiation as observed in RAW 264.7 macrophages. GI 50 values of more than 10 µM were reported in MC3T3-E1 cells with and without 4 Gy radiation (GI 50 > 10 µM).  represents the effects of 4 Gy radiation alone. Cytotoxic effects were evident in DU 145-, MDA-MB-231-, RAW 264.7 cells, and HUVECs exposed to ESE-16. Sensitivity to 4 Gy radiation increased in the RAW 264.7 and HUVEC cells when they were pre-exposed to low-dose ESE-16. MC3T3-E1 cells were spared. Data points depict the means of a minimum of three biological repeats each with an n of at least 3. Standard deviation (SD) indicated by T-bars.

ESE-16-Pre-Sensitization Inhibits Cell Cycle Progression by Inducing G 2 /M Phase Arrest and Promoting Sub-G 1 Phase Accumulation in Cancer and Endothelial Cells
Cell cycle analysis was performed by flow cytometric quantification of PI. Significant increases in the sub-G 1 population were found in DU 145 cells exposed to ESE-16 alone (34.43 ± 1.82%) and the combination treatment (42.33 ± 7.79%) when compared to DMSO (55.30 ± 2.80%) ( Figure 3A). Moreover, the combination treatment yielded a significantly higher sub-G 1 population than 4 Gy radiation alone (12.08 ± 2.25%). The G 2 /M population was significantly higher in the combination treatment (29.45 ± 2.31%) when compared to DMSO (24.56 ± 1.48%) and ESE-16 in isolation (20.29 ± 0.20%). Similar trends were observed in MDA-MB-231 cells. In HUVECs, the DMSO vehicle control indicated a G 1 population of 72.78 ± 1.18% with similar treatment effects as the cancer cells ( Figure 3A). The sub-G 1 and G 2 /M populations were significantly higher in the ESE-16 treatment (9.54 ± 2.24% and 17.45 ± 1.06%, respectively) and the combined therapy (11.39 ± 2.37% and 18.65 ± 1.19%, respectively) when compared to DMSO. Hematoxylin stained negatively-charged nuclear material (purple) and eosin stained positively-charged cytoplasmic contents (pink) ( Figure 3B) [47]. Interphase and the different phases of mitosis (prophase, metaphase, anaphase and telophase) were visualized using light microscopy. The MI was calculated by dividing the total number of cells undergoing mitosis by the total number of cells in the population ( Figure 3C). DU 145 cells exposed to DMSO yielded a MI of 11.03 ± 1.42%, similar to MDA-MB-231 cells (12.73 ± 1.23%). DU 145 cells exposed to ESE-16 or 4 Gy radiation showed no statistical differences in MI, whereas MDA-MB-231 cells exposed to ESE-16 (15.00 ± 0.52%) had a significantly higher MI than DMSO. ESE-16/4 Gy radiation combined yielded significantly higher MI values in MDA-MB-231 cells (17.90 ± 2.52%), but this was not so in DU 145 cells (20.70 ± 6.84%).
The distribution of cells within the mitotic population was investigated ( Figure 3D). DMSO-exposed DU 145 cells showed 35.66 ± 3.79% of cells in prophase, 36.69 ± 2.33% of cells in metaphase, 12.17 ± 3.91% of cells in anaphase, and 7.48 ± 2.84% of cells in telophase. DU 145 cells exposed to the combination treatment demonstrated a significant decrease in prophase cells (27.89 ± 0.75%) and a significant increase in metaphase cells (65.86 ± 1.46%) when compared to DMSO as well as cells exposed to 4 Gy radiation. Anaphase cells (4.71 ± 1.33%) and telophase cells (1.55 ± 1.10%) were also significantly lower in the combination treatment. DMSO-exposed MDA-MB-231 cells showed 46.92 ± 3.07% of cells in prophase, 37.93 ± 5.47% of cells in metaphase, 9.16 ± 4.58% of cells in anaphase, and 5.99 ± 0.75% of cells in telophase. Similar to combination-treated DU 145 cells, MDA-MB-231 cells demonstrated a significant decrease in prophase cells (29.16 ± 3.50%) with a significant increase in metaphase cells (63.80 ± 3.25%) when compared to DMSO as well as cells exposed to 4 Gy radiation. The combination treatment also revealed a significant reduction in telophase cells (1.67 ± 0.54%).

Apoptosis Was Induced in Combination-Treated Cancer Cells, with the Invovlement of CPP32 in MDA-MB-231 Cells
PS is present on the cell's plasma membrane surface and is externalized with the onset of apoptosis. Annexin V conjugated to FITC binds to PS sites on the cell's surface [48]. DU 145 cells exposed to DMSO revealed a viable cell population of 87.82 ± 1.18% ( Figure 4A). The viable cell population was significantly reduced in all treatment conditions when compared to DMSO: 74.09 ± 1.36% in ESE-16-esposed cells, 82.44 ± 1.33% in 4 Gy-irradiated cells, and 66.96 ± 0.25% in the combined therapy which was further significantly lower than the individual treatments. Significantly higher apoptotic populations were observed in all treatment conditions when compared to DMSO: 24.55 ± 1.19% in ESE-16-exposed cells, 15.97 ± 1.01% in 4 Gy-irradiated cells, and 31.19 ± 1.61% in the combination treatment which was significantly higher than either treatment alone. Similar observations were found in MDA-MB-231 cells ( Figure 4A). Cells 2021, 10, x FOR PEER REVIEW 13 of 26  CPP32 (Caspase 3) activity was measured spectrophotometrically to determine whether ESE-16 exposure prior to 4 Gy irradiation involves this protein in the execution of apoptosis [49]. At 24 h post-radiation, significant reductions in CPP32 activity were observed in DU 145 cells exposed to ESE-16 alone (0.93 ± 0.04-fold) and the combined therapy (0.92 ± 0.02-fold). In contrast, MDA-MB-231 cells showed significant increases in CPP32 activity in all treatment conditions: ESE-16 (1.11 ± 0.07-fold), 4 Gy radiation (1.16 ± 0.07-fold), and the combination treatment (1.16 ± 0.09-fold) ( Figure 4B). At 48 h post-radiation, the combined therapy showed no statistical fold-change (FC) in CPP32 activity in DU 145 cells (1.11 ± 0.11-fold), but significantly higher activity in MDA-MB-23 cells (1.10 ± 0.07-fold).

HUVECs Generate Superoxide Ions in Response to the Combination Treatment
Superoxide (O 2 − ) was quantified using flow cytometry in order to evaluate reactive oxygen species (ROS) generation in HUVECs at 24 h post-treatment. Overlay histograms representing O 2 − detection revealed significantly higher geometric means in ESE-16-exposed HUVECs (1.19 ± 0.12-fold) as well as HUVECs exposed to ESE-16 in combination with 4 Gy radiation (1.40 ± 0.27-fold) ( Figure 4C).

Drug Pretreated Cancer Cells Demonstrated Greater Radiation-Induced DNA Damage
Mn are nuclear fragments indicative of double-strand DNA breaks [50]. The average number of Mn per cell as well as the total number of Mn per treatment condition were documented at 2 and 24 h post-radiation in order to determine the extent of DNA damage. DMSO-exposed DU 145 cells showed that the majority of cells had no Mn, 87.00 ± 6.68% and 84.00 ± 3.29% at 2 and 24 h, respectively. This was also observed in MDA-MB-231 cells, 84.67 ± 2.44% and 84.53 ± 2.93% at 2 and 24 h, respectively ( Figure 5A). DU 145 cells exposed to ESE-16 and 4 Gy radiation showed significant decreases in cells without Mn at 2 (39.40  Phosphorylated histone H2A.X (γ-H2A.X) serves as a sensitive biomarker of DNA damage and the extent thereof. H2A.X and γ-H2A.X levels were quantified using flow cytometry [51]. Overlay histograms representing γ-H2A.X expression revealed significantly higher geometric means in combination-treated DU 145-(1.39 ± 0.20-fold) and MDA-MB-231 cells (1.33 ± 0.17-fold) ( Figure 5B). Additionally, combination-treated MDA-MB-231 cells revealed significantly higher γ-H2A.X expression than ESE-16 in isolation (0.86 ± 0.14-fold). In isolation, ESE-16 and radiation did not demonstrate significant differences in γ-H2A.X levels when compared to DMSO.
Fluorescent microscopy was employed to detect γ-H2A.X foci within nuclei. Microscopic results correlated with flow cytometric results in both DU 145 and MDA-MB-231 cells ( Figure 5B). Treatment with ESE-16 and 4 Gy radiation showed comparable results to the etoposide positive control. Increased γ-H2A.X foci within and outside of nuclei were evident and were accompanied by morphological changes in nuclei. ± 0.05-fold) which was also significantly lower than in cells exposed to only 4 Gy radiation over the same time period (1.03 ± 0.05-fold). Thus, the combined therapy prevents the recovery of BMP-7 levels over 24 h when compared to radiation alone. MMP-9 expression remained significantly lower at both 2-and 24-h time periods in DU 145 cells exposed to the combination treatment (0.69 ± 0.13-fold and 0.70 ± 0.10-fold, respectively) ( Figure 6). MDA-MB-231 cells exposed to the combination treatment revealed a significant decrease in MMP-9 expression at 2 h (0.59 ± 0.14-fold) with a further decrease at 24 h (0.23 ± 0.14-fold). Furthermore, combination-treated MDA-MB-231 cells showed significantly lower MMP-9 expression at 24 h when compared to the ESE-16 experimental control (0.62 ± 0.12-fold).

Pre-Osteoclasts Retentained Theirr Differentiation Capacity after Treatment with the Comvination Therepy, and Actin Ring Formation Was Preserved in Mature Osteoclast
TRAP is highly expressed in mature osteoclasts and serves as a histochemical marker for osteoclastogenesis [46]. The effect of the combination treatment on TRAP in RAW 264.7 macrophages either before or after differentiation was investigated to determine the ability of pre-osteoclasts to maintain their differentiation potential, and to determine if the treatment is toxic to mature osteoclasts. RAW 264.7 macrophages exposed to ESE-16 and 4 Gy radiation prior to differentiation showed very slight (but significant) increases in TRAP activity in the 4 Gy radiation-(1.02 ± 1.50 × 10 −3 -fold) and combination (1.03 ± 0.02-fold) treatment conditions ( Figure 7A). In contrast, RAW 264.7 macrophages treated after differentiation into mature osteoclasts showed significant decreases in TRAP activity in all treatment conditions: ESE-16 (0.93 ± 0.05-fold), 4 Gy radiation (0.91 ± 0.04-fold) and the combined therapy (0.92 ± 0.07-fold). Mature multinucleated osteoclasts (≥3 nuclei) were visualized with the TRAP stain (red) and hematoxylin (purple) ( Figure 7A). No statistically significant changes were documented in osteoclast numbers when treatment was performed on pre-osteoclasts. In contrast, small but significant reductions in osteoclast numbers were recorded in all treatment conditions when treatment was performed on fully-differentiated mature osteoclasts. Fold-decreases of 0.75 ± 0.10, 0.84 ± 0.05, and 0.79 ± 0.03 were observed in the ESE-16-, 4 Gy radiation, and combination treatments, respectively. Thus, mature osteoclasts display increased susceptibility to treatment compared to pre-osteoclasts, but the combination does not appear to be more toxic than the experimental controls.

Discussion
MTAs disrupt microtubules dynamics and halt mitosis at metaphase due to the inability of cells to successfully pass the spindle assembly checkpoint (SAC), resulting in the induction of apoptosis [33]. Sulfamoylated estrone derivatives of 2ME2, such as ESE-16, exhibit improved bioavailability and enhanced potency when compared to the parent compound [36]. Previously, a sister compound of ESE-16 namely 2-ethyl-3-O-sulfamoyl- Actin rings are responsible for isolating bone matrix from the extracellular environment in order for its degradation during bone resorption [54]. Actin ring formation is thus an indirect indicator of optimal resorptive activity. Since statistically significant effects were observed on treated mature osteoclasts in TRAP experiments, the effects thereof were investigated on actin ring formation. Fluorescent microscopy was employed to visualize actin rings (green) and cell nuclei (blue) ( Figure 7B). Large multinucleated osteoclasts with intact actin rings were observed in the RANKL+ and DMSO controls. Compared to the DMSO control, the individual therapies, ESE-16, and 4 Gy radiation alone showed fewer actin rings that were smaller in size. These effects were exacerbated in the combination treatment; however, actin rings remained intact, suggesting that the osteoclasts retained their resorptive capability.

Discussion
MTAs disrupt microtubules dynamics and halt mitosis at metaphase due to the inability of cells to successfully pass the spindle assembly checkpoint (SAC), resulting in the induction of apoptosis [33]. Sulfamoylated estrone derivatives of 2ME2, such as ESE-16, exhibit improved bioavailability and enhanced potency when compared to the parent compound [36]. Previously, a sister compound of ESE-16 namely 2-ethyl-3-O-sulfamoyl-estra-1,3,5(10),15-tetraen-17-ol (ESE-15-ol) demonstrated radiosensitizing properties in lung carcinoma (A549) and MCF-7 cells. Colony formation assays indicated decreased long-term cell survival and regeneration [55]. This proof of concept study aimed to examine the cell signaling involved in the induction of the cell death by a low-dose drug-radiation combination treatment, and thus earlier timepoints post-exposure were studied. Additionally, components of a bone metastatic lesion were screened for their responses to this treatment in order to identify potential aspects worth examining in greater detail in future.
In this study, ESE-16 displayed cytotoxicity at nanomolar concentration in DU 145 and MDA-MB-231-cancer cells, as well as RAW 264.7 macrophages and HUVECs. Response to radiation was enhanced by pre-exposure to the drug. MC3T3-E1 pre-osteoblastic bone cells displayed relative resistance to ESE-16 alone as well as in combination with 4 Gy radiation.
In previously published research, a significant increase in the mitotic and sub-G 1 population with a corresponding reduction in the G 1 population was reported on cell cycle analysis of MCF-7 cells pre-sensitized with ESE-15-ol prior to 6 Gy irradiation [55]. In this study, DU-145, MDA-MB-231 cells, and HUVECs exposed to ESE-16 and 4 Gy radiation in combination showed similar results. In DU 145 and MDA-MB-231 cells, the combination treatment was more cytotoxic than radiation alone, and caused greater mitotic inhibition than ESE-16 alone. Microscopy findings supported these data, although MDA-MB-231 cells treated with ESE-16 alone or in combination with 4 Gy radiation yielded more a pronounced effect on the MI than DU 145 cells. Both cell lines revealed a significant increase in metaphase cells when compared to the radiation treatment control. The mitotic block displayed in the combination treated cells, together with the decreased cell density demonstrated, was most likely augmented by the microtubule abrogative properties of ESE-16 and the mitotic catastrophe caused by the radiation [56].
The combined effects of ESE-16 and 4 Gy radiation on cell survival were evident by a significant decrease in DU 145 and MDA-MB-231 cell viability and a significant increase in apoptosis. These effects surpassed those generated in the ESE-16 and radiation controls, suggesting enhanced or augmented potency. These effects may not yet be striking, as early timepoints were investigated in an effort to understand the signaling and the cellular interplay of the two modalities. These results correlate with a previous study investigating the radiosensitizing effects of ESE-15-ol [55]. A significant reduction in cell viability with a concurrent increase in apoptosis was observed in A549 and MCF-7 cells pre-exposed to low-dose ESE-15-ol prior to 6 Gy irradiation. Long-term survival was significantly decreased in the combination treatment as determined by clonogenic studies.
Caspase 3 (CPP32) activation is a landmark event of caspase-dependent apoptosis [57,58]. MDA-MB-231 and HeLa cells exposed to 0.2 µM of ESE-16 yielded a small but significant increase in CPP32 activity after a 24-h-exposure, an effect that was amplified by an increased dose [38,39,59]. In this study, the positive method control (paclitaxel) confirmed that the DU 145 and MDA-MB-231 cell lines used express caspase 3. Combinationtreated MDA-MB-231 cells revealed involvement of CPP32 in the cell death signaling which declined from 24 to 48 h. DU 145 cells did not show an increased CPP32 activation in response to treatment at the studied time periods and at the low treatment doses. It is possible that late apoptotic events involving caspase 3 activity have not yet occurred at these time periods or that executioner caspases 6 and/or -8 may alternatively be involved in concluding apoptosis [60]. Temporal activation studies would aim to give clarity to this hypothesis, as well as investigation into caspase-independent death pathways.
ROS, such as O 2 − , are chemically reactive oxygen-containing molecules that are generated as a by-product of normal oxygen metabolism and radiation exposure. Since ROS interact with intracellular biomolecules, an imbalance in ROS generating and scavenging systems has the potential to damage DNA, and induce intrinsic apoptotic pathways [61]. Oesophageal carcinoma (SNO), MCF-7, and MDA-MB-231 cells exposed to ESE-16 generated increased O 2 − levels in a temporal response to the drug exposure [38,42]. In this study, O 2 − levels in HUVECs increased significantly when treated with ESE-16 alone or in combination with 4 Gy radiation.
Mn represent residual chromosomal damage resulting from mis-or non-repaired double-strand DNA breaks derived from radiation exposure, and thus serve as an indication of chromosomal radiosensitivity [62,63]. At 2 and 24 h, DU 145 and MDA-MB-231 cells treated with ESE-16 prior to 4 Gy irradiation revealed a significant prevalence of cells containing between 1 and 4 Mn when compared to the ESE-16 treatment control. Moreover, the combined therapy yielded total Mn numbers that were significantly greater than ESE-16 alone but not statistically different from radiation alone. However, higher Mn numbers per cell that were observed in the combination treatment suggest a greater extent of DNA damage than caused by radiation alone. ESE-16 itself does not appear to have genotoxicity. ESE-16 may facilitate Mn formation when combined with radiation due to its effects on the microtubule dynamics which result abrogative cell division processes (including a prophase and metaphase block), and could delay DNA repair by altering DNA damage and repair signaling due to compromised protein shuttling [64].
γ-H2A.X foci form at phosphorylated sites and are directly proportional to the extent of DNA damage [65,66]. 2ME2 in combination with 6 Gy X-radiation increased the formation of γ-H2A.X foci in radioresistant MDA-MB-435R cells, more so than radiation exclusively [31]. In this study, histone H2A.X phosphorylation increased significantly in combination-treated DU 145 and MDA-MB-231 cells when compared to DMSO, an observation absent in the individual treatment controls. Since Kim et al. observed higher γ-H2A.X expression in prostatospheres than in adherent LNCaP cells following exposure to 10 Gy ionizing radiation [61], these results would do well to be repeated in a 3-D spheroid model. Fluorescent γ-H2A.X foci were evident in all treatment conditions with pan-nuclear staining possibly indicating apoptotic bodies or DNA fragments [67].
The osteogenic protein BMP-7 and endopeptidase MMP-9 are upregulated in both breast and prostate cancer bone metastases. They facilitate mesenchymal-epithelial transition, promote invasive and migratory properties of tumor cells to bone sites and support distant tumor survival within the bone microenvironment [12,68]. BMP-7 and MMP-9 expression was inhibited in DU 145 and MDA-MB-231 cells exposed to ESE-16 and 4 Gy radiation. Expression declined further at 24 h in MDA-MB-231 cells, whereas DU 145 cells indicated similar inhibitory effects at both time periods. The combination treatment delayed recovery of BMP-7 expression in MDA-MB-231 cells, as seen in the radiation control. This may be suggestive of impaired signaling between the primary tumor cells to the bone sites, as well as impaired survival of metastatic tumor cells within the bone microenvironment. BMP-7 expression is higher in breast cancer than in prostate cancer [68]. This may account for the ability of DU 145 cells to stabilise BMP-7 levels more rapidly. A trending decline in MMP-9 in MDA-MB-231 cells may limit tumor cell migration and adhesion to bone as well as decrease osteoclast differentiation and activation [12,69]. Future experiments will include quantification of MMP-9 activity via zymology in cellular co-culture protocols.
HIF-1α promotes the adaptive responses of tumor cells to their hypoxic microenvironment through the downstream transcriptional activation of genes regulating tumor survival and progression. These genes express proteins that facilitate cellular oxygen and nutrient transport such as erythropoietin, VEGF-A, and transferrin [70][71][72]. HIF-1α messenger ribonucleic acid (mRNA) levels, protein expression, and relative density were inhibited in 2ME2-treated MDA-MB-435R cells [31]. In this study, ESE-16 and combination-treated HUVECs displayed significant reductions in HIF-1α expression at 2 and 24 h. Furthermore, HIF-1α expression in the combination treatment condition was significantly lower than observed in the radiation treatment control, attributing this suppression solely to the drug effect.
TRAP is an iron-containing enzyme with high expression in mature osteoclasts and essential involvement in bone resorption [73]. Fully-differentiated mature osteoclasts treated with ESE-16 and 4 Gy radiation were more sensitive than pre-osteoclasts treated prior to differentiation. This suggests that the surviving population of treated pre-osteoclasts maintain their ability to differentiate into mature functional osteoclasts. However, treated mature osteoclasts displayed a significant reduction in osteoclast numbers with a concurrent reduction in TRAP activity.
At the site of bone resorption, mature osteoclasts polarise and attach to the bone surface via actin-rich podosomes which form circular actin rings that isolate bone matrix. Actin ring formation in mature osteoclasts is an important indirect indicator of correct polarization and resorptive activity [54]. 2ME2 demonstrated preservation of bone density in ovariectomized rats, suggesting inhibition of bone resorption [74]. Since TRAP experiments rendered mature osteoclasts more sensitive than pre-osteoclasts, this necessitated further investigation on treated mature osteoclast by examining actin ring formation. Decreased osteoclast densities were accompanied by actin rings that were smaller in size. However, actin ring morphology was retained. These observations suggest that bone resorptive activity may be compromised, but not completely lost. Since treatment had less of a cytotoxic effect on the MC3T3-E1 pre-osteoblasts, the combination treatment could perhaps shift the balance of bone remodelling towards bone formation, and in this way could elicit bone-protective effects. This however would need to be studied in a more holistic cellular model using primary osteoblast cultures and bone marrow-derived macrophages in breast and prostate cancer conditioned media. MDA-MB-231 cells do not express RANKL; however, high tumor necrosis factor-α (TNF-α) secretion by these cells stimulates downstream p38 signaling pathways involved in osteoclast differentiation and inflammatory bone loss [75][76][77].
Thus, ESE-16 may radiosensitize cancer cells by pausing the cell cycle at the sensitive G 2 /M phase, increasing ROS production, causing an abrogative autophagic response, and delaying the DNA damage repair by impeding the cytoplasmic-nuclear translocation of DNA damage repair proteins, amongst others [41,44]. A limitation of this deconstructed monolayer cellular model is the lack of the metastatic microenvironment consisting of various intricate cellular interactions. As an example, soluble breast cancer factors are responsible for the transformation of human mammary fibroblasts into cancer-associated fibroblasts which promote tumor growth and have increased invasive and migratory properties [78]. The results of this study provide a foundation for future studies in which the bone microenvironment may be simulated, including co-culture systems and spheroid models. Establishment of osteolytic metastatic breast cancer cell lines, such as those described by Han et al. may be instrumental in analyzing the effect of the combination treatment on the metastatic mechanisms involved in bone lesions [79].

Conclusions
This study explored the individual responses of the representative components of bone metastasis in which ESE-16 pre-sensitization enhanced the effects of radiation. Overall, the ESE-16/radiation combination treatment induced a G 2 /M phase arrest (metaphase block) in both cancer and endothelial cells along with apoptosis induction and ROS generation. Early and late Mn analyses indicated sustained radiation-induced nuclear damage in cancer cells with phosphorylation of histone H2A.X (γ-H2A.X foci) indicating double-strand DNA breaks. Inhibition of metastatic signaling through the downregulation of BMP-7 and MMP-9 expression may interrupt bone-tumor communication. Additionally, inhibition of angiogenic signaling through the downregulation of the hypoxic responder HIF-1α may reduce tumor growth. Cytotoxic effects were observed in pre-osteoclasts whilst preosteoblasts were spared. Preservation of TRAP activity and actin ring formation suggest maintained pre-osteoclastic differentiation capacity and possible retention of osteoclastic resorptive activity This combined treatment strategy shows promise in the treatment of bone metastases by targeting tumor and endothelial cells whilst sparing bone cells. Thus, the use of non-lethal doses of ESE-16 together with a lower dose of radiation may reduce the occurrence of local and systemic adverse effects. Future research will focus on in vitro 3-dimensional co-culture models to imitate the tumor micromilieu in bone metastasis.

Supplementary Materials:
The following are available online at https://www.mdpi.com/article/ 10.3390/cells10082105/s1, Supplementary Data: Figure S1: Experimental set-up using cell cycle analysis. Figure S2: Experimental set-up using Annexin V-FITC apoptosis detection. Table S1: ESE-16 concentrations and radiation doses used in all further combination studies.