1. Introduction
Osteosarcoma is a high-grade primary mesenchymal tumor characterized by spindle cells depositing an immature osteoid matrix [
1]. To date, osteosarcoma is the most frequent primary malignancy of bone in children and the most frequent primary malignancy in adolescents apart from leukemia and lymphoma [
2,
3]. Surgical excision is often effective only for patients with low-grade tumors [
4]. For patients with high-grade tumors, other therapeutic methods, such as chemotherapy and radiotherapy, must also be employed [
5]. Chemotherapy used in osteosarcoma protocols remains essentially unchanged since the introduction of high-dose methotrexate, doxorubicin, and cisplatin in the late 1970s [
6,
7,
8]. The five-year overall survival has remained approximately 60% over the last five decades; nevertheless, the overall survival of patients with metastatic osteosarcoma is <20% [
9]. Multiple efforts to improve therapeutic efficacy have not identified more effective or less toxic regimens, despite intensifying treatment or modulating the immune response [
7,
10,
11,
12]. Therefore, new therapeutic approaches are urgently needed.
Induced differentiation of transformed cells into mature phenotypes has proven to be an effective strategy in the treatment of several types of human malignancies [
13,
14], and derivatives of vitamin A, retinoids, are some of the most frequently used inducers of differentiation [
15,
16,
17,
18]. The molecular mechanism of retinoid signaling is based on their binding to members of the nuclear receptor family, retinoic acid receptor (RAR) and retinoid X receptor (RXR), which subsequently form homodimers or heterodimers, bind to the DNA, and influence transcription directly, or they can interact with other transcription factors. In addition to their nuclear transcriptional effects, retinoids are able to rapidly and transiently activate several kinase signaling pathways [
19].
Despite the many benefits of retinoids as anticancer compounds, their usage in clinical protocols is still limited because of their short intracellular availability, clinically significant toxicity, and the occurrence of resistance [
20]. Therefore, efforts have been made to include retinoids in combined treatment with other drugs that may enhance or prolong their antineoplastic effects. Combinations of all-
trans retinoic acid (ATRA) with several natural compounds, kinase inhibitors, chemotherapeutics, and proteasome inhibitors have demonstrated additive or synergistic effects [
21]. Our research group described the enhancement of the antineoplastic effect of ATRA caused by inhibition of its catabolism using LOX/COX inhibitors (caffeic acid and celecoxib) in neuroblastoma, medulloblastoma, and osteosarcoma cell lines [
22,
23,
24,
25]. The benefits of combined treatment in the therapy of several solid tumors have also been confirmed for retinoic acid and other differentiation inducers, such as calcitriol [
26,
27,
28].
Calcitriol (1α,25(OH)
2 vitamin D
3) is the most biologically active form of vitamin D
3 [
29]. It is mainly synthesized endogenously via UVB radiation of human skin followed by stepwise hydroxylation in the liver and kidney or can be obtained by exogenous dietary intake [
30,
31]. In animal cells, calcitriol binds to the nuclear vitamin D receptor (VDR), which is subsequently transported to the nucleus, where it forms dimers. The dimer complex acts as a transcription factor that can either activate or suppress mRNA expression after binding to the vitamin D responsive element in the promotor region of several target genes that are primarily involved in the calcium homeostasis of cell differentiation, in bone formation, resorption, and mineralization, and in the maintenance of neuromuscular function [
32]. The recent meta-analysis suggests that calcitriol and its precursor calcidiol (25(OH) vitamin D
3) could act as chemopreventive agents [
33]. The correlations between low serum levels of calcidiol and increased mortality of patients with colorectal cancer [
34], prostate cancer [
35], breast cancer [
36] and melanoma [
37] have also been reported. To date, several studies on the antineoplastic effects of calcitriol in osteosarcoma have been published [
38,
39,
40,
41,
42,
43]. Nevertheless, the dose-dependent response to calcitriol and calcidiol in different osteosarcoma cell lines is still not well defined, and the mechanisms involving the inhibition of proliferation and differentiation induction remain unclear [
44].
In the present study, we focused on the possible effects of calcitriol and calcidiol alone or in combination with ATRA in patient-derived osteosarcoma cell lines, with special regard to the mechanism of interaction between calcitriol and ATRA.
3. Discussion
In the present study, we described the responsiveness of seven human osteosarcoma cell lines to two forms of vitamin D3 (calcitriol and calcidiol) and to their combinations with the known differentiation inducer ATRA. The Saos-2 established cell line and six patient-derived cell lines were used for experiments.
According to the results from the MTT assay, the Saos-2 established cell line showed only minimal sensitivity to the treatment with calcitriol or calcidiol alone. Although the antiproliferative effect of calcitriol in Saos-2 cell line has already been reported [
45,
46], another study showed that neither 100 nM calcidiol nor 10 nM calcitriol inhibited proliferative activity in Saos-2 cells after 96 h of treatment [
40]. We suspected that this lack of visible inhibition might be due to the early endpoint (96-h), which was not long enough for calcitriol to mediate its downstream action. Therefore, the treatment was extended to 168 h, but no changes in cell proliferation were visible. As the insensitivity of Saos-2 cells was observed in terms of proliferation activity only (i.e., Saos-2 cells were sensitive in terms of induced differentiation) in our experiments, we assume that these inconsistencies may also be caused by different methods of evaluation of the proliferation activity.
In the combination treatments, only calcitriol was able to significantly enhance the inhibitory effect of ATRA. Similarly, we observed that the mRNA level of BGLAP, an osteogenic differentiation marker, was highest after combined treatment with calcitriol and ATRA during the entire analyzed period.
The sensitivity of the patient-derived cell lines to differentiation inducers was indeed specific to each cell line. We realized that the increased sensitivity to all differentiation inducers, including calcidiol, in the OSA-13 cell line could be caused by the low differentiation stage of those cells. OSA-13 was previously described as a tumorigenic cell line with elevated expression of the transcriptional factor SOX-2 [
47].
Variability in the responsiveness of cell lines could also be related to differences in endogenous levels of respective nuclear receptors for calcitriol and ATRA, as both compounds function as ligands for the respective receptors and subsequently change gene expression [
48,
49]. We hypothesized that more sensitive cell lines express higher endogenous levels of relevant receptors for these drugs. However, this hypothesis was not confirmed. In general, the expression of respective nuclear receptors in untreated cell lines did not correspond to the inhibition effect of the drugs.
Subsequently, we focused on the evaluation of nuclear receptor expression after 24 h of differentiation inducer treatment and observed that 1 µM ATRA was able to regulate VDR expression. This phenomenon has already been observed in mouse and rat osteosarcoma cell lines [
50,
51,
52,
53]. Changes in VDR levels caused by ATRA have already been described in monocytic leukemia cell lines. On the one hand, the majority of research suggests that treatment with ATRA alone is sufficient for VDR regulation [
54,
55,
56,
57,
58]. On the other hand, one study suggested that only the combined treatment of ATRA and calcitriol effectively increased VDR protein levels but not
VDR mRNA expression in the THP-1 human monocytic leukemia cell line [
59]. In this study, ATRA as a single agent was not able to regulate VDR at the mRNA or protein level [
59].
Our results are consistent with the findings described above. Twenty-four hours of treatment with 1 µM ATRA caused changes in
VDR mRNA levels and VDR protein levels in selected osteosarcoma cell lines. Upregulation or downregulation of VDR depended on the cell line. It was described that there is no
RARE in the
VDR promoter, which suggested that ATRA could not regulate
VDR directly [
60]. Therefore, it is assumed that retinoids can regulate
VDR transcription indirectly using regulatory elements that cooperate with the
VDR promoter. Moreover, a study on myeloid leukemia cell lines showed that the most important isoform of RAR involved in the regulation of
VDR transcription is RARα. In the absence of ligands, RARα led to transcriptional repression of the
VDR gene in this cell type [
57].
In accordance with these studies, we focused on the RARα isoform and its agonist ATRA. For better interpretation, we compared the two most different osteosarcoma cell lines—the Saos-2 reference cell line, which had the lowest level of RARα and the highest level of VDR, and the OSA-08 cell line, which had the highest level of RARα and the lowest level of VDR. In the Saos-2 cell line, downregulated expression of VDR was observed after RARα activation by ATRA at both the mRNA and protein levels. In contrast, upregulated expression of VDR was detected in the OSA-08 cell line after ATRA treatment.
Moreover, the combined effect of ATRA and calcitriol was the most effective in the OSA-08 and OSA-13 cell lines, which had high levels of RARα. These data correlate with the hypothesis that unbound RARα acts as a transcriptional repressor of
VDR [
57]. We assume that there is a mechanism involving a change in the RARα conformation after ATRA binding that releases the repression of
VDR by RARα. After the repression is overcome, cells start to express higher levels of VDR and calcitriol, thus inducing a stronger response. This response even enhances the antineoplastic effect of ATRA, so the combination is more effective than the effect of each drug alone. According to this hypothesis, we expected to see high sensitivity to calcitriol in cell lines with low endogenous expression of RARα, but our experimental data on cell proliferation did not confirm this idea: Saos-2, i.e., the cell line with the lowest endogenous level of RARα, did not respond to calcitriol at any of the used concentrations. In this case, we must take into account that another mechanism of resistance to vitamin D
3, i.e., an overexpression of VDRE-BP, could be activated in Saos-2 cells [
61].
To summarize, our results proved that combination treatment with calcitriol and ATRA showed an enhanced antiproliferative effect compared with the effect of those drugs alone in the majority of tested cell lines. Furthermore, this study provides the first evidence that ATRA treatment influences VDR expression in human osteosarcoma cells in vitro. More specifically, ATRA upregulated VDR expression at the mRNA and protein levels in cell lines with high endogenous levels of RARα and low endogenous levels of VDR; only these cell lines were the most sensitive to the combination treatment. In general, the results suggest that the levels of RARα and VDR in osteosarcoma cells could potentially be used as predictors of possible synergy between calcitriol and ATRA.
4. Materials and Methods
4.1. Cell Culture
The Saos-2 established cell line (No. HTB-85) was purchased from the American Type Culture Collection (Manassas, VA, USA). Other cell lines were derived from tumor samples obtained during diagnostic biopsies from patients suffering from osteosarcomas. The samples were processed in our laboratory as previously described [
62]. The OSA-02, OSA-03, OSA-05, OSA-08, and OSA-13 cell lines were already used and described in our previous studies [
47,
63,
64]. The OSA-09 cell line was derived from the sample of conventional osteosarcoma taken from a 22-year-old patient. The Research Ethics Committee of the School of Medicine (Masaryk University, Brno, Czech Republic) approved the study protocol, and a written statement of informed consent was obtained from each patient or his/her legal guardian prior to participation in this study.
Cells were grown in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% (Saos-2 cells) or 20% (OSA-02, OSA-03, OSA-05, OSA-08, OSA-09, and OSA-13 cells) fetal bovine serum, 100 IU/mL penicillin, 100 mg/mL streptomycin, and 2 mM glutamine (all purchased from GE Healthcare Europe GmbH, Freiburg, Germany). Cell culture was performed under standard conditions at 37 °C in a humidified atmosphere containing 5% CO
2. Patient-derived cell lines at passages 10–25 were used for the experiments (
Supplement 2).
4.2. Chemicals
Calcitriol (Sigma-Aldrich, St. Louis, MO, USA) and calcidiol (Sigma-Aldrich) were prepared as stock solutions at a concentration of 1 mM in absolute ethanol (Penta, Prague, Czech Republic) and stored at −20 °C. ATRA (Sigma-Aldrich) was prepared as a stock solution at a concentration of 100 mM in DMSO (Sigma-Aldrich) and stored at −20 °C under light-free conditions. All three stock solutions were freshly diluted in cell culture medium for each use.
4.3. Treatment
For proliferation tests, 96-well plates were seeded with 5 × 103 cells per well (Saos-2 cells) or 2 × 103 cells per well (OSA-02, OSA-03, OSA-05, OSA-06, OSA-09, and OSA-13 cells) in 200 μL of complete DMEM. Cells were allowed to adhere overnight. Subsequently, the medium was removed, and fresh medium containing the appropriate concentrations of drugs alone or in combination was added. Cells were treated with five concentrations of calcitriol (10 pM, 100 pM, 1 nM, 10 nM, and 100 nM), five concentrations of calcidiol (100 pM, 1 nM, 10 nM, 100 nM, and 1 μM), and one concentration of ATRA (1 μM). The plates were incubated under standard conditions for 3 or 7 days.
To prepare samples for immunoblotting and PCR analyses, cells were seeded onto Petri dishes and allowed to adhere overnight. The medium was removed and replaced with fresh medium containing 10 nM calcitriol, 100 nM calcidiol, and/or 1 μM ATRA. For immunoblotting and qPCR, cells were harvested after 24 h of treatment, and for semiquantitative RT-PCR, cells were harvested after 1, 3, 5, 7, and 9 days of treatment.
In all experiments, untreated cells were used as controls. In addition, we compared the proliferation activity of untreated cells and cells treated with vehicle (DMSO/ethanol) only and found no significant difference.
4.4. Cell Viability
Cell viability was evaluated using the MTT assay, which was performed as previously described [
25]. Briefly, the plates with 0.5 mg/mL 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl- tetrazolium bromide (MTT) (Sigma-Aldrich) were incubated at 37 °C for 3 hours. Formazan crystals were dissolved in 200 μL of DMSO. The absorbance at 570 nm was measured with a reference absorbance at 620 nm using a Sunrise Absorbance Reader (Tecan, Männedorf, Switzerland). Each experiment was performed in triplicate. The results obtained were expressed as a percentage of untreated controls.
4.5. RT-PCR
The expression of osteogenic differentiation markers was evaluated using semiquantitative RT-PCR. The protocol included standard procedures that were previously described [
15]. The primers for genes of interest are listed in
Table 2. The optical density of bands was quantified using ImageJ software, and the data were normalized to
HSP90AB1 expression. Each experiment was performed in triplicate.
The relative expression levels of selected nuclear receptors were studied using RT-qPCR. Total RNA was extracted and reverse transcribed into cDNA in the same manner as described previously [
25]. RT-qPCR was carried out in 10 μL using the KAPA SYBR
® FAST qPCR Kit (Kapa Biosystems, Wilmington, MA, USA) and analyzed using the 7500 Fast Real-Time PCR System and 7500 Software v. 2.0.6 (both Life Technologies, Carlsbad, CA, USA). Changes in the transcript levels were calculated using Cq values standardized to a housekeeping gene (
GAPDH) used as an endogenous reference gene control. The established Saos-2 cell line served as the arbitrary calibrator. The primers used for genes of interest are provided in
Table 5. Each experiment was performed in triplicate.
4.6. Immunoblotting
Cells were lysed in LB1 buffer (50 mM Hepes-KOH, pH 7.5, 140 mM NaCl, 1 mM EDTA, 10% glycerol, 0.5% NP-40, 0.25% Triton X-100), and the total protein amount was subsequently measured by the DC Protein Arrays Reagents Package (Bio-Rad Laboratories, Munich, Germany) according to the manufacturer’s instructions. Total proteins (10 μg) were loaded onto 10% polyacrylamide gels, electrophoresed, and blotted on a polyvinylidene difluoride membrane (Bio-Rad Laboratories). The membranes were blocked with 5% nonfat dry milk in PBS with 0.1% Tween-20 (Sigma-Aldrich) and incubated with primary antibodies overnight. The next day, membranes were incubated with secondary antibodies at room temperature (RT) for 1 hour. All antibodies used for immunoblotting are listed in
Table 6. ECL-Plus detection was performed according to the manufacturer’s instructions (GE Healthcare). The optical density of bands was quantified using ImageJ software, and the data were normalized to loading control GAPDH. Each experiment was performed in triplicate.
4.7. Statistics
Quantitative data were statistically evaluated using SPSS Statistics software (version 25.0, IBM, New York, USA). Data obtained in the MTT assay were analyzed by one-way ANOVA, followed by the Scheffé post hoc test: *
p < 0.05 and **
p < 0.001 were considered statistically significant. Analysis of possible interactions of compounds included in this study was performed using the Bliss independence model [
65]. Data obtained using PCR and immunoblotting were analyzed with a one-sample
t-test (two-tailed): *
p < 0.05 was considered statistically significant.