In vitro study of combined cilengitide and radiation treatment in breast cancer cell lines

Background Brain metastasis from breast cancer poses a major clinical challenge. Integrins play a role in regulating adhesion, growth, motility, and survival, and have been shown to be critical for metastatic growth in the brain in preclinical models. Cilengitide, an αvβ3/αvβ5 integrin inhibitor, has previously been studied as an anti-cancer drug in various tumor types. Previous studies have shown additive effects of cilengitide and radiation in lung cancer and glioblastoma cell lines. The ability of cilengitide to enhance the effects of radiation was examined preclinically in the setting of breast cancer to assess its possible efficacy in the setting of brain metastasis from breast cancer. Methods Our panel of breast cells was composed of four cell lines: T-47D (ER/PR+, Her2-, luminal A), MCF-7 (ER/PR+, Her2-, luminal A), MDA-MB-231 (TNBC, basal B), MDA-MB-468 (TNBC, basal A). The presence of cilengitide targets, β3 and β5 integrin, was first determined. Cell detachment was determined by cell counting, cell proliferation was determined by MTS proliferation assay, and apoptosis was measured by Annexin V staining and flow cytometry. The efficacy of cilengitide treatment alone was analyzed, followed by assessment of combined cilengitide and radiation treatment. Integrin β3 knockdown was performed, followed by cilengitide and radiation treatment to test for incomplete target inhibition by cilengitide, in high β3 expressing cells. Results We observed that all cell lines examined expressed both β3 and β5 integrin and that cilengitide was able to induce cell detachment and reduced proliferation in our panel. Annexin V assays revealed that a portion of these effects was due to cilengitide-induced apoptosis. Combined treatment with cilengitide and radiation served to further reduce proliferation compared to either treatment alone. Following β3 integrin knockdown, radiosensitization in combination with cilengitide was observed in a previously non-responsive cell line (MDA-MB-231). Clonogenic assays suggested little radiosensitization effects of cilengitide. Conclusions Cilengitide appears to enhance radiation response in preclinical models of breast cancer. These data suggest that the combination of radiation therapy and cilengitide may prove to be effective where radiation is utilized for the treatment of gross disease in breast cancer, such as in the setting of brain metastasis.


Background
Brain metastasis from breast cancer occurs in approximately 5% of patients overall, and in 10-16% of patients with metastatic disease [1]. Incidence is thought to be on the rise, as systemic therapy advances lead to better local tumor control and improved survival. Current treatments for these metastases include whole brain radiotherapy, surgery, stereotactic radiosurgery, and chemotherapy [2]. Outcomes for these patients are poor with median survivals ranging from 3.4-25.3 months based on the Graded Prognostic Assessment (GPA) [3]. Radiation is commonly used in the treatment of brain metastasis from breast cancer, where there can be gross disease present in the brain. Given that a significant percentage of these patients succumb to their metastatic disease and demonstrate local progression of their disease in the brain, we sought to investigate agents that may demonstrate additive or synergistic effects with radiation in the setting of breast cancer.
Integrins play a role in regulating cell-extracellular matrix interactions as well as cell signaling pathways that regulate adhesion, growth, motility, and survival [4]. Integrins are expressed on endothelial cells, and play an important role in angiogenesis [5], but have also been identified on a number of cancer cell types [6][7][8]. αVβ3 integrins specifically have been identified to play a direct role in tumor cell growth as well as invasion and metastasis [9,10]. αVβ3 integrins were shown to be critical for metastatic growth of breast cancer cells in the brain [11]. Therefore, we further examined targeting integrin signaling in combination with radiation.
Cilengitide is a cyclic RGD containing pentapeptide that targets αVβ3 and αVβ5 integrins [12]. This inhibitor has been shown to block glioma cell growth via cell detachment and induction of apoptosis in an in vitro model [13]. In vivo, cilengitide has been shown to inhibit metastatic bone colonization by the breast cancer cell line MDA-MB-231 [14].
In the context of radiation therapy, combination treatment with cilengitide has been shown to radiosensitize lung cancer cell lines [15], with lung cancer representing the tumor site with the highest incidence of metastasis to the brain [1]. Combination therapy of cilengitide and radioimmunotherapy with an L6 antigen targeting antibody conjugated with the beta-emitter 90 Y has shown to improve outcomes of primary breast tumors in a xenograft model of breast cancer [16]. The combination of cilengitide and external beam radiotherapy has yet to be studied in the context of breast cancer.
We therefore set out to determine if a combination therapy of cilengitide and radiation could be of benefit for breast cancer brain metastases patients. To this end we tested the effect of cilengitide in combination with radiation in a panel of breast cancer cell lines, among them cell lines that have previously been used to study brain metastases from breast cancer.

Flow cytometry -apoptosis
Apoptosis assays were performed using the Alexa Fluor® 488 Annexin V/Dead Cell Apoptosis Kit (Life Technologies) according to the manufacturer's instructions. Briefly, cells were plated and allowed to attach overnight. The following morning, cells were treated with indicated cilengitide doses and allowed to incubate for 48 hours. Apoptotic induction controls were treated with 16.6 uM cisplatin 24 hours before staining. After 48 hours, cells were harvested (floating and attached cells) and stained according to manufacturer's instructions. Single stained and unstained controls were included. Flow cytometry was performed on a BD Facs LSR II flow cytometer under guidance from flow cytometry core staff. Samples were analyzed using Flow Jo software.

Clonogenic assay
All cell lines were trypsinized and counted, followed by dilution to appropriate levels. Cells were seeded in triplicate in 2 ml growth media into 6-well tissue culture dishes and allowed to attach overnight. The following morning, the indicated doses of cilengitide were added to the cells. Irradiation was performed at the indicated doses 1 hour after drug treatment. Cells were then maintained at 37°C in the appropriate atmosphere for 10 days to 1 month until appropriate sized colonies had formed. Once colonies were visible, cells were stained in 0.5% crystal violet in methanol for 2 hours. Colonies were counted with the use of a dissecting microscope with a cutoff of 50 cells. Dose enhancement ratios are reported for 37% cell survival.

Cell counting
Cells were plated at 2 × 10 5 cells/well and allowed to attach. Cell counting was performed following 1 hr treatment with the indicated cilengitide doses. Detached cells were removed and wells were washed with PBS. Remaining attached cells were then trypsinized and diluted 1:2 with trypan blue. Live cells were counted in duplicate using the Nexelcom Cellometer Auto T4 cell counter (Nexelcom Bioscience). Remaining cell viability was at least 95%.

MTS assay
Proliferation assays were performed using the CellTiter Aqueous One Solution Cell Proliferation Assay (Promega) according to modified manufacturer's instructions. 2 × 10 4 cells/well were plated in 500 ml growth medium in triplicate. Cells were treated at the indicated dose of cilengitide on day 0. Control readings were measured on day 0 to control for plating inaccuracies and cell metabolism differences. The CellTiter solution was diluted 1:6 and 240 ul/ well was added. Plates were incubated for 3 hrs at 37°C, followed by removal of 100 ul of MTS solution. Absorbance at 490 nm was obtained using a Mithras LB 940 plate reader (Berthold Technologies). Treated cells were left for 96 hours and absorbencies were obtained as above. ITGB3 knockdowns were performed using ON-TARGET Plus SMARTpool Human ITGB3 siRNA and Control Non-Targeting siRNA. Cells were treated with 25 nM siRNA for 48 hours before plating. Day 4 values were controlled to Day 0 values within each cell line. Values were also controlled to untreated cell values and reported as % Cell Viability compared to untreated cells. For the ITGB3 knockdown experiments, normalization was performed against the untreated control cells for each group (Non-Targeting cells or ITGB3 knockdown cells respectively).

Results
Cilengitide target expression in breast cell lines β3 and β5 integrins, found as dimers with αV integrins, are the targets for the inhibitor cilengitide. αVβ3 integrin is thought to represent the primary target, while αVβ5 is an alternate target. We examined the breast cancer cell lines T-47D, MCF-7, MDA-MB-231, and MDA-MB-468 for β3 and β5 integrin expression by western blot. As shown in Figure

Effects of cilengitide alone Cell detachment
Once expression of the target of cilengitide was verified, we assessed the effects of cilengitide treatment as a single therapy in our panel of cell lines. Previous studies have shown that cilengitide causes cell detachment in culture [13,15]. The role of integrins in cell adhesion to uncoated plastic dishes is probably related to serum vitronectin adhering to the plastic [17]. We wanted to examine this cellular detachment in our panel of cell lines. Briefly cells were treated with cilengitide (500 nM, 5 uM, and 20 uM) for 1 hour. Detached cells were then washed away with PBS, and remaining attached cells were trypsinized and counted. We observed that in most of our cell lines there was a dose dependent effect of cilengitide on cell detachment. At the highest dose tested (20 uM) T-47D cells showed the most marked response, with almost complete detachment of cells following 1 hour of treatment ( Figure 2). MDA-MB-468 cells showed little to no cell detachment at this early timepoint, even at this highest dose. Both of the other cell lines tested showed a moderate response to the 20 uM dose, losing about 50% of plated cells (Figure 2). Most of these cells, however, are likely to be merely detached, as opposed to dead. Cells treated for 1 hour with 20 uM cilengitide to cause detachment, were then washed with PBS and re-plated into a new cell culture dish. At least a part of these washed cells were able to attach and grow normally (data not shown). Cells left in cilengitide up to 96 hours remained detached (data not shown).

Growth inhibition
The same trends discussed above were observed in our panel of cell lines in a longer term proliferation assay. Breast cancer cell lines were plated, allowed to attach overnight, and then treated the following morning with varying doses of cilengitide (1 uM, 5 uM, 10 uM, and 20 uM). The cells were then allowed to grow for 96 hours in the presence of cilengitide and cell viability was determined. As shown in Figure  response when compared to the previous early cell detachment assay, showing only 25% reduction in cell proliferation after 96 hours of cilengitide exposure compared to controls as opposed to 50% of cells detached after 1 hour ( Figures 2C  and 3C). On the other hand MCF7 cells showed a more noticeable response, with nearly an 80% reduction in cell proliferation compared to the 50% of cells lost after 1 hr of treatment ( Figures 2B and 3B). Overall these results are consistent with the early cell detachment results. Cilengitide appears to have a significant effect, on both cell attachment and cell growth, in our breast cancer cell lines.

Cilengitide induces apoptosis
To examine the mechanism by which cilengitide reduces viability in our cells, we performed an apoptosis assay at an intermediate timepoint (

Effects of cilengitide in combination with radiation
After characterizing the response of our cell lines to cilengitide as a single therapy, we sought out to determine if cilengitide could serve as a radiosensitizer. First, we performed a proliferation assay (MTS) as discussed earlier ( Clonogenic assays showed only very minor increases in radiosensitivity following cilengitide treatment in some of the cell lines tested (Figure 6). These results indicate that cilengitide is enhancing the effects of radiation in our panel of cells, and that there may be an additive effect for this combination treatment.
It was observed that the least responsive cell lines in this study were those with the highest levels of β3 and lowest levels of β5 integrin target. The association of high expression levels of integrin β3 and poor response to cilengitide in our small cell line panel suggests that integrin β3 is the relevant cilengitide target. One possibility for these results is that this target is not completely inhibited at the cilengitide doses used. Since integrin β5 expression was associated with responsive cell lines, we determined that this target was likely being adequately inhibited by the cilengitide doses studied. To examine if these cells were unresponsive to cilengitide treatment due to incomplete β3 blockade at tested treatment doses, β3 integrin was knocked down in the MDA-MB-231 cell line. Approximately 50% knockdown was achieved in these cell lines ( Figure 7A). 48 hours following knockdown, cells were plated and treated as in the above experiments. Figure 7B shows that cells with reduced β3 integrin show enhanced response to radiation and cilengitide combination treatment, while previously they were minimally responsive to cilengitide and radiation treatment ( Figure 5).

Discussion
Brain metastases from breast cancer occur in 10-16% of patients who eventually develop metastatic disease [1]. The advent of brain metastases often is a life altering event for breast cancer patients and not uncommonly impacts quality of life and cognitive functioning. Outcomes for patients with brain metastases are poor with median survival times ranging from 3.4-25.3 months depending on GPA [3]. There is an urgent need for novel therapeutic approaches to improve outcomes in this patient population. In this study we present our findings suggesting cilengitide treatment can induce cellular detachment and apoptosis, and reduce proliferation in a panel of breast cancer cell lines. While this is, to our knowledge, the first report to describe cilengitide effects in combination with radiation in a panel of breast cancer cell lines our findings are consistent with previous reports describing induction of apoptosis, reduced proliferation, and cellular detachment after cilengitide treatment in glioma [13] and lung cancer [15] models.
The primary target of cilengitide is αvβ3 integrin. Integrin expression on tumor cells plays a role in activating many of the major cell survival pathways including ERK and PI3K signaling [18]. These pathways are known to lead to increased tumor cell proliferation, migration, and resistance to apoptosis [19]. Furthermore inhibition of those pathways can result in radiosensitizing effects in many cancer cell lines [20]. Therefore, we speculate that these pathways could mediate the treatment effects in breast cancer cell lines as well, and it is not surprising that the integrin inhibitor cilengitide has anti-tumor activity in lung and glioma models or in breast cancer models as here presented.
The magnitude of cilengitide effects on breast cancer cell lines, however, was fairly heterogeneous among the cell lines tested. Given that only four cell lines were tested no definitive conclusions can be drawn at this time as to which breast cancer brain metastases patient subpopulations might derive the most benefit from cilengitide containing treatment regimens. It appeared that cell lines that are ER/PR positive and Her2 negative or cell lines with lower expression levels of integrin β3 had the most significant response to cilengitide treatment in this study. Experimentally reducing the expression of integrin β3 in a cell line with high integrin β3 expression (MDA-MB-231) resulted in increased efficacy of cilengitide and radiation. Given that integrin β3 knockdown can sensitize cells to the tested cilengitide doses, it appears that this target is not completely inhibited in cell lines with high expression of integrin β3. While it is still possible that integrin β5 inhibition is responsible for the enhanced response seen in the T-47D or MCF-7 cell lines, these results indicate that at least a portion of the cilengitide resistance in our other cell lines is due to the high levels of integrin β3 expression observed. While the results of our proliferation assays have shown clear statistical benefit of combination therapy compared to either therapy alone in some breast cell lines, our clonogenic data shows little to no formal radiosensitization effect. Our results indicate a more significant benefit of combined radiation and cilengitide treatment at earlier timepoints compared to the long term results of the clonogenic assays. This difference could well be due to the different timepoints tested. However, the performed normalization of the radiated arm to the cilengitide alone treated cells likely contributes to this observed difference as well. Our data however is similar to reports using glioma models [21] where modest radiosensitizing effects of cilengitide were found. Previously it has been reported that cilengitide and radiation have additive effects in glioma cell lines [22]. Our data indicate a similar additive nature of combination cilengitide and radiation therapy in breast cancer cell lines, rather than a synergistic or supra-additive effect. These results are more limited in scope when compared to more significant radiosensitizing effects of cilengitide reported in NSCLC models [15]. The underlying reasons for the differences in potency of cilengitide as a radiosensitizer in different cancers remain to be identified and is subject to further studies. Despite the differences in efficacy seen comparing our assays, combined cilengitide and radiation treatment appeared to be beneficial for most of the tested breast cancer cell lines. In summary, our data suggest cilengitide treatment has anti-tumor activity in breast cancer cell line models, and that combination treatment with radiation appears to affect cells more than either treatment alone. Given that some of the cell lines used in our study have previously been used as breast cancer brain metastases models [23], and that cilengitide potentially can achieve therapeutically relevant concentrations in the brain, we conclude that combined cilengitide and radiation treatment could be a promising therapeutic strategy for a subset of breast cancer brain metastases patients.

Conclusions
Combined cilengitide and radiation therapy appears to be more efficacious than either treatment alone in breast cancer cell lines. This study, along with previous work highlights the promise of cilengitide therapy in combination with radiation. Based on the ability of cilengitide to accumulate in the brain combination treatment of cilengitide and radiation could be beneficial clinically for a subset of breast cancer brain metastases patients.