- Top of page
- Materials and Methods
- Semi-quantitative RT-PCR
- Competing Interests
Bromoacetoxy-calcidiol (B3CD), a pro-apoptotic and cytotoxic agent in neuroblastoma (NB) cell lines, displayed therapeutic potential in vivo as an anticancer drug in a NB xenograft mouse model. Tumors of all animals treated intraperitoneally with B3CD went into regression within 10–30 days of treatment, while tumors in control animals grew aggressively. The response mechanisms of NB cells to B3CD in vitro were studied and included differential targeting of cell cycle key regulators p21 and cyclin D1 on the transcriptional and expression level leading to arrest in G0/G1 phase. In contrast to the effect in ovarian cancer cells, B3CD-induced cell death in SMS-KCNR NB cells was only marginally mediated by the p38 MAPK signaling pathway. Signaling induced by exogenous recombinant EGF leads to a partial restoration of the negative effects of B3CD on SMS-KCNR cell proliferation and survival. Upon combinational treatment of SMS-KCNR cells with B3CD and recombinant EGF, the EGF receptor (EGF-R) was highly activated. We suggest future studies to include analysis of the effects of B3CD in combination therapy with pharmacological inhibitors of cell cycle regulators or with EGF-R-targeting inhibitors, -toxins or -antibodies in vitro and their translation into in vivo models of tumor development.
Neuroblastoma (NB) is a rare cancer of the peripheral sympathetic nervous system. Peripheral neuroblastic tumors (pNTs) range from benign ganglioneuroma to stroma-rich ganglioneuroblastoma with well-differentiated neuroblastic cells to highly malignant NB (1). Three distinct cell types have been isolated from NB cell lines: N-type cells with properties of embryonic sympathoadrenoblasts, S-type cells resembling non-neuronal Schwannian, glial, melanoblastic precursors, and I-type stem cells that can differentiate into either N- or S-type cells (2,3). About 500 new cases of NB are diagnosed in the United States each year (4). The majority of NB cases occur in children below the age of five and NB account for 7–10% of all childhood cancers. In the majority of patients older than 1 year of age, the disease is fatal. Multimodal treatment methods include surgery, radiation therapy, chemotherapy, autologous stem-cell transplantation (5–7) either alone or in combination, depending on the location and biological characteristics of the cancer cells, stage, and the risk group to which the patient belongs. However, more than 50% of children with high-risk disease relapse, because of drug-resistant residual disease (8–10). Eradication of refractory microscopic disease remains one of the most significant challenges in the treatment of the high-risk NB and innovative treatments need to be designed.
Past studies led to the development of bromoacetoxy-calcidiol (B3CD, Figure 1), a bromoacetoxy-ester derivative of calcidiol, which exerted potent selective antiproliferative effects on prostate cancer cells (11–13) and NB cell lines (14). Calcidiol is the natural precursor to calcitriol/vitamin D3, is found abundantly in serum, and is biologically inactive both in terms of binding to the vitamin D receptor (VDR) and in terms of transcription regulation (15). B3CD through indirect approaches was suggested to interact directly with the VDR receptor and in prostate cancer cells the authors hypothesize that this drug exerts cellular effects via the VDR signaling pathway (11). Bromoacetoxy analogs such as B3CD generally display an improved pharmacologic profile, exert less toxicity, and greater stability compared to their parent compounds (16,17). Previous in vitro studies showed that B3CD at concentrations as low as 1.0 μm displayed strong growth-inhibitory effects in prostate cancer cell lines while other cancer cells such as breast cancer cells or primary keratinocytes were significantly less affected (12,13). Previous studies on various neuroblastoma cell lines revealed high cytotoxicity of B3CD at 1 μm and antiproliferative effects with IC50 concentrations as low as 30–100 nm (14). Cell death of NB cells upon treatment with B3CD is mediated by the intrinsic signaling pathway of apoptosis (14), whereas for prostate cancer cells, in addition to the intrinsic pathway, B3CD-induced apoptosis is mediated by the extrinsic pathway (11). In NB cells (SMS-KCNR), the cytotoxic response to B3CD is correlated with suppression of Akt-mediated pro-survival signaling as well as with suppression of the oncogenic transcription factor MYCN (14), which is overexpressed in more than 65% of human NB (18). In ovarian cancer cells (SKOV-3), B3CD-induced cell death is directly mediated by p38 MAPK function (19), which is essential for EGF-dependent ovarian cancer invasiveness (20). Interestingly, NB cells lines express a variety of EGF receptors, and EGF can stimulate the proliferation of NB cell lines in vitro (21) and induce expression of pro-survival factors including p38 (22).
Figure 1. VDR Expression in Neuroblastoma cell lines after treatment with Calcidiol derivative B3CD. (A) Structure of B3CD and precursor Calcidiol. (B) Vitamin D receptor expression. SMSK-CNR or SK-N-SH cells were treated with 1 or 3 μm B3CD for 48 h. Western blot analysis of cell lysates was carried out as described (Material and Methods) using primary antibodies against VDR. As an internal standard for equal loading, the blots were probed with an anti-GAPDH antibody.
Download figure to PowerPoint
The objective of the present study was to investigate the therapeutic potential of B3CD to treat NB in vivo in a NB xenograft animal model. Because B3CD was postulated to exert cellular effects via the VDR signaling pathway (11) we analyzed the expression change of the VDR receptor upon B3CD treatment of NB cell lines SMS-KCNR and SK-N-SH in correlation to the cytotoxicity exerted by the drug. We addressed the hypothesis that B3CD-induced cell death, similar to ovarian cancer cells (19), may be mediated by p38 signaling and might be altered by the growth-stimulating effects of growth factor EGF. Because B3CD has previously been reported to affect cell cycle progression in SMS-KCNR cells (14), we studied the expression profile of several cell cycle regulators upon BC3D treatment.
- Top of page
- Materials and Methods
- Semi-quantitative RT-PCR
- Competing Interests
Previous work has shown that B3CD displayed strong growth-inhibitory effects with respect to both proliferation and viability in certain prostate cancer cell lines (11) while other cancer cells such as MCF-7 (breast cancer) or primary keratinocytes were less affected (12,13). Thus, B3CD in contrast to the parent compound calcitriol/vitamin D3, which is highly calcemic in vivo (30–32), was proposed to be a potential anticancer drug specific for certain cancer types. However, in a previous study, we observed that B3CD, even though highly cytotoxic to defined cell lines including certain ovarian cancer cells (SKOV-3) did not reveal specificity with respect to tumor origin (other ovarian cancer cell lines were less responsive) (19). When B3CD was used as a drug in an ovarian cancer animal model (human SKOV-3 cell derived xenografts in nude mice), mixed results were obtained. We observed that the majority of B3CD-treated mice displayed delayed tumor growth or full tumor regression while in a few B3CD-treated mice tumor growth accelerated (19). Accordingly, for the treatment of ovarian cancer, we proposed further development of non-calcemic bromoacetoxy derivatives of calcitriol/vitaminD3 as potential anticancer therapeutics.
In contrast to ovarian cancer cell lines, a variety of NB cell lines studied revealed a consistent cytotoxic response to B3CD treatment (14). B3CD, depending on the cell line and concentration tested, displayed up to fivefold greater cytotoxicity than the parent compound calcidiol. The structural difference between B3CD and calcidiol is the presence of a bromoacetate functionality, which must be correlated to the efficacy of this drug in NB cells. This observation led us to investigate the potential anticancer activity of B3CD in a NB xenograft model for the present study. We chose (NU/NU) mice originated from the NIH, which lack a thymus, are unable to produce T cells and are used for xenograft and syngeneic tumor studies. Choice of dosage of the drug (non-toxic concentration of 150 μg/kg body weight) was based on previous studies. A recent systemic study in CD-1 mice showed that B3CD did not raise serum calcium nor exhibit toxicity (166 μg/kg, repeated intraperitoneal administration unlike other synthetic vitamin D3 derivatives (12), and references therein). In general, bromoacetic acid, a possible metabolite after B3CD administration in clinically relevant doses is unlikely to reach toxic concentrations; the LD50 of bromoacetic acid in male rats is 88 mg/kg (33), which is several 100-fold higher than concentrations of B3CD used in our in vivo studies.
B3CD treatment of NB tumor cell derived xenografts in mice revealed a chemotherapeutic effect within 6 days of treatment. From day 6 forward the average tumor in treated animals was reduced in diameter when compared to treatment start. In contrast, tumors in control mice grew aggressively. Because of increased tumor burden in the controls, the study was ended at 30 days. By study, end tumors in untreated mice were 3.75 times larger in diameter than tumors in B3CD-treated mice. Within this time frame complete responses (full tumor regression) to B3CD were not observed. However, in the treatment group, tumor size stabilized within 22 days of treatment at ∼3 mm diameter. In summary, B3CD displayed anticancer activities in this NB xenograft animal model. We did not observe ulcerations around NB cell derived tumor sites or acceleration of tumor growth, unlike our previous study using ovarian cancer cell xenografts in the same mouse strain (19). Given the fact that B3CD (i) did not cause adverse effects and (ii) revealed activity against neuroblastoma xenografts, we propose to continue the evaluation of this bromoacetoxy ester derivative of calcidiol in neuroblastoma cancer model systems. We also initiated studies to develop other non-calcemic vitamin D derivatives as potential anticancer agents (34). In addition, we suggest studies on the effects of B3CD in combination therapy with other anticancer drugs, antibodies, or cytokines in tumor cells in vitro and their translation into in vivo models of tumor development.
Like various other cytotoxic agents, B3CD displays cell cycle regulatory effects in NB cells at the IC50 as well as at subcytotoxic concentrations (14). After B3CD treatment, the subdiploidal apoptotic population increases and this correlates with the onset of apoptotic signaling and DNA fragmentation (14). With respect to the cycling cells, B3CD causes an increase, or depending on the concentration, an arrest of the G0/G1 population along with decrease in cells in S-phase and G2/M phase.
In the present report, we analyzed the direct effect of B3CD on regulators of cell cycle progression such as cyclin D1, or p21 and p27 (both are regulators of cyclin D-dependent kinases; CDK) in SMS-KCNR and SH-SY5Y NB cells. In these cell lines, B3CD at a concentration of 1 μm, which mildly suppresses viability of NB cells, targets two key cell cycle regulators, p21 and cyclin D1. It acts by upregulating p21, downregulating cyclin D1 expression, and downregulating the transcription of cyclin D1 but not of p21. Drug targeting of cell cycle checkpoints and key regulators, which are frequently altered in human cancer, such as cyclin D1 and p21, has been suggested as an alternative approach to anticancer therapies (28,35,36). We also observed that at the cytotoxic dosage of 3 μm B3CD, p27 expression and transcription were downregulated. It has been reported that depending on the experimental parameters, either p27 upregulation or p27 downregulation can be associated with a G1 arrest in NB cells (37). P27 sequestration by cyclinD-cdk complexes is a key factor in establishing a balance between cell proliferation and cycle arrest even though p27 knockout animals are relatively free of malignancy (36; and references therein). Given the multifaceted role displayed by p27 and the fact that p27 expression in NB cells changes only upon high dosage of B3CD, we suggest that p27 regulation is not a major regulator of specific cell cycle regulation exerted by B3CD. In contrast, cyclin D1 and p21 are modulated by B3CD at lower dosage. Cyclin D1 downregulation and p21 induction leading to inhibition of cyclin D-dependent kinases is a prerequisite for the arrest of cells in G1 phase in general (28,36) as well as in NB cell lines. However, modulation of a regulator alone depending on the experimental setting and cell system does not necessarily lead to cell cycle arrest as was shown for NB cells where p21 (as well as p53) induction lead to apoptosis, yet G1 cell cycle arrest was attenuated (38). Similarly, inhibition of CDK activity by itself does not necessarily lead to the arrest of cycling cancer cells (28). In contrast, B3CD interferes with the progression of cells through G1 phase by regulation of cyclin D1 and p21 expression. Therefore, the use of drugs such as B3CD alone or in combination with pharmacological CDK inhibitors may specifically prevent the progression of NB.
In a previous study in ovarian cancer cells, we observed a dramatic suppression of the cytotoxicity of B3CD by interfering with the activity of p38 MAPK while inhibition of other MAPKs such as Erk 1, Erk2, JNK 1, 2 and 3, Jun/JNK, or MEK did not significantly alter B3CD-mediated cell death (19). P38 and other MAPKs such as JNK, MEK, Erk1/2 mediate signaling pathways in cancer and control cell lines responding to inflammatory cytokines, UV light, cytotoxic drugs, and diverse other pro-apoptotic stimuli (39–41). Activation of p38 generally is a pro-apoptotic trigger and is a key determinant for drug-induced apoptosis, such as cisplatin in ovarian cancer cells (42). Similarly, sustained activation of the p38 and JNK MAPK pathways by different drugs (e.g. Fenretinide) initiated cell death of NB cell lines (43) and certain apoptotic stimuli in NB cells act mainly via the p38 MAPK pathway (44). B3CD caused an upregulation of the activity of this pro-apoptotic signaling factor in SMS-KCNR NB cells (19). However, in SMS-KCNR cells, p38 inhibitors could only partially counteract the effect of B3CD and restore cell viability by only 5–10%. This observation suggests that the p38 MAPK signaling pathway in NB cells, unlike that of ovarian cancer cells, does not play a pivotal role in the response to B3CD.
We observed a stimulating effect of exogenously added recombinant EGF on SMS-KCNR NB cell proliferation. It has been reported that EGF can stimulate the growth of a variety of NB cell lines in vitro along with activation of both mitogen-activated protein kinase (MAPK) and phosphoinositide 3-kinase (PI3K)/AKT pathways (21). It is known that NB primary tumors and cells lines express a variety of growth factor receptors including epidermal growth factor receptors (EGF-R; HER1-4) (21). Moreover, resistance of NB to standard drugs, such as cisplatin, has been directly linked to enhanced levels of EGF-R expression. Such cells are sensitive to treatment with specific toxins and antibodies targeting EGF-R (45). Consequently, an alternative treatment option for NB might be a combination treatment with EGF-R-targeting agents as well as standard (e.g. cisplatin) or newly developed drugs such as B3CD. We analyzed whether the cytotoxic effect of B3CD on NB cells might be altered by exposure to exogenous recombinant EGF and whether these treatments change the expression levels of the EGF-R.
Recombinant EGF, added at 20 nm concentration, prevented almost completely the antiproliferative effect of 1 μm B3CD on these NB cells. Similarly, EGF reversed the mild antiproliferative effect of calcitriol/vitamin D3 on SMS-KCNR cells. In addition, recombinant EGF at 40 nm could partially counteract the cytotoxic effect of B3CD on this NB cell line. Therefore, we hypothesize that signaling induced by EGF leads to a restoration of both the negative effects of subcytotoxic concentrations of B3CD on cell cycle progression and of cytotoxic concentrations on NB cell death. EGF signaling is primarily induced by EGF receptors (HER1-4). Upon combination treatment of SMS-KCNR cells with B3CD and recombinant EGF, EGF receptors were highly activated. Apparently, access to exogenous EGF allowed NB to counteract B3CD-induced cell death by an increase in EGF-R signaling as an effort to survive drug treatment. Apart from specific toxins or antibodies clinically established EGF-R-specific tyrosine kinase inhibitors such as Gefitinib (also known as Iressa or ZD1839) could be evaluated in their effect on NB in combinational treatment with B3CD. Even though effective in vitro concentrations cannot be clinically reached, Gefitinib also displayed chemosensitizing effects when used with other drugs (e.g. topotecan, vincristine) in NB cell lines (46). In summary, we postulate that the chemotherapeutic properties of B3CD might be enhanced by co-treatment with EGF-R-targeting agents (e.g. toxins, antibodies, tyrosine kinase inhibitors) and suggest further animal models to substantiate the findings.