Abstract
Background/Aim: This study evaluated the effect of haematogenous administration of acridine orange (AO) alone and in combination with zoledronate (ZOL) on bone metastases. Materials and Methods: E0771 cells (1.0×105 cells/10 μl) were injected directly into the right femur of female mice. The mice were divided into five groups according to treatment (drugs and irradiation) and were reared and sacrificed after 6 weeks. Micro-computed tomography (μCT) was performed to calculate the destruction rate of the femur bone. We measured tumour weight and volume at sacrifice and performed terminal deoxynucleotidyl transferase dUTP Nick-End Labelling staining of tumours. Results: At 4 weeks, the bone destruction rate was lower in the AO+ZOL group than in the radiation group. At 6 weeks, the AO+ZOL group had a lower bone destruction rate than the control and radiation groups; the ZOL group had a lower rate than the radiation group. The AO and AO+ZOL groups had suppressed tumour weight and volume compared to the control and radiation groups. The number of extraosseous apoptotic cells was higher in the AO+ZOL group than in all other groups except the AO group. Conclusion: In a model of local bone metastasis of breast cancer, haematogenous administration of AO reduced tumour size and more so when combined with ZOL.
The number of patients with cancer increases every year, and by 2025, more than 20 million new cases are expected to occur annually, mainly in developing countries (1). Currently, Japan hosts a substantial number of patients with cancer, which has been increasing since 2000 (2). Additionally, bone metastases occur at a high rate in patients with advanced cancer. Therefore, as the number of patients with cancer increases, that of patients with bone metastasis is also expected to increase. Bone metastases often occur in the spine; however, they can appear anywhere in the body, and many cases consist of multiple metastases (3). Additionally, bone metastases occur most frequently in breast cancer, with 65-75% of patients with breast cancer reported to have bone metastases (4, 5). Radiation therapy and surgery are used to treat the paralysis and pain caused by bone metastases; however, they are only localized treatments. Currently, the established systemic therapy consists of administration of bone resorption inhibitors, such as bisphosphonates and denosumab (6). Nevertheless, the therapeutic efficacy of these agents is limited, and new systemic therapies for patients with metastatic bone tumours are needed.
The first step in the development of a novel therapy is to study it in an established animal model. To date, metastatic bone tumour models in mice have generally been created by systemic haematogenous administration of tumour cells (7). However, this model is limited in its ability to produce metastatic bone tumours at the target site, hindering constant evaluation. Nevertheless, locally administered bone metastasis models using knockout mice have been reported, but they have not been developed because of the choice of growth environment and infection. Therefore, we previously created a localized bone metastasis model that could easily and reliably create bone metastases at a fixed location and proved its usefulness using zoledronate (ZOL) (8).
As an agent for systemic therapy, we focused on acridine orange (AO), a photosensitive fluorescent dye. AO is made from acridine, an organic molecule derived from coal, and has been used in various stains, including that for endoscopic diagnosis of gastric cancer (9). AO is an acoustic-, light-, and radiation-sensitive substance with a specific affinity for cancer cells. Moreover, the pH of AO, a weakly basic compound, has been reported to be related to and promote the intracellular pH of cancer cells, which are often acidic (10). Furthermore, AO can promote apoptosis of cancer cells upon external stimulation. Photodynamic therapy and local administration of AO utilizing these properties have been reported to reduce the local recurrence rate of sarcomas (11-18). Furthermore, intravascular administration of AO in mouse models has been reported to reduce local tumour size (19) and inhibit lung metastasis (20). However, the effect of AO on carcinomas and their bone metastases has not yet been investigated.
In this study, we confirmed the efficacy of haematogenous administration of AO in a model of local bone metastasis from breast cancer. The efficacy of AO in combination with ZOL, which has been reported to be effective in bone metastasis of carcinoma, was also examined.
Materials and Methods
Cell cultures. E0771 (CH3 Biosystems LLC, Buffalo, NY, USA) cells were cultured in Dulbecco’s modified Eagle’s medium (Sigma-Aldrich, St. Louis, MO, USA) supplemented with 10% foetal bovine serum (Mediatech, Manassas, VA, USA) and 100 μg/ml kanamycin sulphate (Meiji Seika Pharma, Tokyo, Japan). The cultures were maintained in a humidified atmosphere of 5% CO2 at 37°C (21). The cells were verified to be mycoplasma-free before being injected into the mice using polymerase chain reaction (ICLAS Monitoring Center, Kawasaki, Japan).
The cells were diluted in phosphate-buffered saline (PBS) so that the final number of cells was 1.0×105 cells/10 μl. The survival rate of the tumour cells was evaluated using the trypan blue dye exclusion method with a haemocytometer (Kayagaki, Tokyo, Japan) under an optical microscope (Olympus BH-210, Tokyo, Japan, ×400).
Creation of the model of local bone metastasis in the femur. The protocols for the animal experiments described in this paper were previously approved by the Animal Research Committee, Akita University School of Medicine (Protocol #: a-1-3163), and conducted according to the ARRIVE guidelines.
In this study, we used a model of local femoral bone metastasis in C57BL/6 mice previously described by our research group. Four-week-old female C57BL/6 mice (Charles River Laboratory Inc., Kanagawa, Japan) were housed in a specific pathogen-free environment. The mice were anesthetized, and E0771 cells were administered topically. In detail, a combination anaesthetic was prepared with 0.3 mg/kg of medetomidine, 4.0 mg/kg of midazolam, and 5.0 mg/kg of butorphanol. The anaesthetic was administered via subcutaneous injection to obtain a sufficient depth. We made a median incision in the knee of each mouse, and the patella was flipped laterally to expose the femoral condyle. We created a bone socket in the femur using a 26-gauge needle. Finally, E0771 cells (1.0×105/10 μl) were suspended in 10 μl of PBS injected using a Hamilton syringe (8).
Protocol. Mice were divided into five groups according to the use of therapeutic intervention or irradiation (Figure 1) as follows: (i) control group without treatment intervention (Con), (ii) those treated with ZOL (ZOL), (iii) those treated with AO and irradiation (AO), (iv) those treated with irradiation only (Rad), and (v) those treated with AO, ZOL, and irradiation (AO+ZOL). The concentration of AO was set at 1 mg/kg single dose in accordance with previous reports (22). AO was administered intravascularly through the tail vein under anaesthesia using the same method used to create the bone metastasis model 3 weeks after inoculation when the appearance of local bone tumours could be confirmed (19). Two hours after AO administration, irradiation stimulation was performed using CP-160 (Faxitron, Tucson, AZ, USA) to activate the AO. The irradiation dose was 5 Gy whole body irradiation as in previous reports to activate AO (23). The dose of ZOL was 100 μg/kg, which is equivalent to a 4 mg infusion for the treatment of bone metastases in humans (24). ZOL was administered 2 weeks after tumour cell administration, in accordance with a previous report (25). All five groups of mice were sacrificed after 6 weeks of rearing.
Experimental groups and schedule. Mice were divided into five groups: (1) control group (Con, n=10); (2) zoledronate-treated group (ZOL, n=10); (3) acridine orange-treated group (AO, n=10); (4) radiation exposure group (Rad, n=10); and (5) acridine orange + zoledronate-treated group (AO+ZOL, n=10). All groups received a local injection of tumour cells at 4 weeks of age. Each group was treated with drugs or radiation, respectively, and sacrificed after 6 weeks.
Evaluation. The development of bone metastasis was monitored by micro-computed tomography (μCT) analysis using micro-focus X-ray CT CosmoScan GX II (Rigaku Corporation, Tokyo, Japan). Three-dimensional digital images were reconstructed using bone analysis software (Rigaku Corporation). The mice were monitored at 3, 4, and 6 weeks post-injection, and the degree of bone destruction was calculated. To calculate the rate of bone destruction using μCT, we first measured the length of the femur from the femoral head to the femoral condyle in the sagittal section. The axial section was used to identify the location where cortical bone destruction was partially observed, and the sagittal section was used to confirm the length of bone destruction. The following formula was used: femur length with the appearance of bone destruction/femur length ×10 (8). The appearance of systemic metastases was confirmed by photographing the entire body with μCT after sacrifice. Tumour volume and weight were measured by removing the right thigh and tumour as a single mass, and a calliper was used to measure the volume. Tumour volume was calculated by the following formula: (2× short diameter × long diameter) ×0.5 (6). After the mice were sacrificed, bone volume and bone surface were measured using μCT at 25-35 mm proximally from the femoral condyle where the tumour cells were injected. Histological evaluation was performed with TUNEL staining to assess tumour cell apoptosis (26). Tissues were fixed in neutral formalin, embedded in paraffin, and sectioned at 5 μm. The TACS2 TdT-DAB In Situ Apoptosis Detection Kit (R&D Systems®, Inc., Minneapolis, MN, USA) was used for TUNEL staining. Tumour cells were quantitatively evaluated using an all-in-one BZ-X800 fluorescence microscope (KEYENCE, Osaka, Japan). The number of apoptotic cells was measured in three fields of view at 400× magnification inside and outside the femur; these values were averaged.
The protocol for animal experiments described in this paper was approved in advance by the Animal Experiment Committee of Akita University School of Medicine, and all subsequent animal experiments were conducted in accordance with the “Animal Experiment Guidelines” of Akita University.
Statistical analysis. Data are expressed as mean±standard deviation, and comparisons between groups were analysed through ANOVA (R Development Core Team, R Foundation for Statistical Computing, Vienna, Austria) (Table I). Statistical significance was set at p<0.05.
Summary of comparison of the five groups.
Results
The only occurrence of distant metastasis was a cervical lymph node metastasis in one mouse in the Con group. Tumour weight was significantly lower in the AO group than in the Con, ZOL, and Rad groups (p<0.05). The AO+ZOL group also showed a significantly lower tumour weight than the ZOL group (p<0.05). Regarding tumour volume, the AO and AO+ZOL groups resulted in significantly smaller tumour volumes than the Rad group (p<0.05). At 4 weeks after tumour cell injection, the bone destruction rate was significantly lower in the AO+ZOL group than in the Rad group (p<0.05). At 6 weeks after injection, the bone destruction rate was significantly lower in the AO+ZOL group than in the Con and Rad groups, and in the ZOL group than in the Rad group (p<0.05) (Figure 2). Although no significant difference was observed in the number of apoptotic cells inside the bone by TUNEL staining, the number of apoptotic cells outside the bone was significantly higher in the AO group than in the ZOL group (p<0.05). Finally, the AO+ZOL group showed significantly more extraosseous apoptotic cells than the other three groups (p<0.05) (Figure 3).
Femur 3D reconstruction image by μCT 6 weeks after tumour cell administration. The zoledronate (ZOL) group (b) and the acridine orange (AO)+ZOL group (d) show that cortical bone is preserved, and bone destruction is suppressed compared to the Con (a) and AO groups (c).
Histological sections stained with TUNEL outside the femur injected with tumour cells. The AO (c) and AO+ZOL groups (d) had more TUNEL-positive and apoptotic cells in extraosseous tumours than the Con (a) and ZOL groups (b). 200× magnification.
Discussion
In this study, haematogenous administration of AO enhanced apoptosis of tumour cells and significantly reduced tumour weight and volume. Although AO is known to show a specific affinity for cancer cells (10, 11), it selectively accumulated in tumour cells and led to apoptosis in a local metastatic bone tumour model. This study employed breast cancer cells, suggesting that haematogenous administration of AO may have an effect on carcinomas as well as bone and soft tissue malignancies. Although the effects of AO on cancer osteosarcoma models are thought to be both apoptotic and necrotic (19), it was thought that AO may also induce apoptosis in breast cancer cells. However, haematogenous administration of AO alone did not suppress bone destruction, which is thought to be caused by cancer, nor did it significantly increase apoptotic cells in bone. This result suggests that AO may act differently in extra-bone and intra-bone cells and therefore may not have shown a strong inhibitory effect on bone destruction. Further evaluation of the intraosseous and extraosseous mechanisms is needed to further improve the efficacy of AO.
In contrast, ZOL administration significantly reduced bone destruction in the local metastatic bone tumour model. ZOL has been previously reported to have an effect on bone metastases through osteoclast- and oxidative stress-mediated pathways (27) and showed a significant inhibitory effect on bone destruction in the present study. However, this study did not show any effect on shrinkage of the tumour or promotion of apoptosis. Past reports have shown the effect of ZOL not only on bone metastases but also on the primary tumour (6, 27). In such cases, the induction of apoptosis of cancer cells by Ras protein inhibition and the angiogenesis inhibitory pathway were shown to be the mechanism of action. Meanwhile, other studies have reported that ZOL promotes apoptosis of intraosseous tumours, but not of extraosseous tumours (26). In the present study, no significant difference was observed between the two, but ZOL treatment showed a trend towards more apoptosis within the bone. Although various factors such as differences in tumour cell lines may be involved, a more detailed evaluation of intra- and extra-bone tissues is necessary.
Sites where cancer has metastasized to the bone tend to have an acidic environment due to tumour cells. The tumour cells secrete acid because of the presence of Na+/H+ exchange transport, HCO3- transport, proton-lactate transport, and the proton pump (28). Furthermore, cancer cells produce cytokines that promote osteoclast differentiation and activity, and activated osteoclasts secrete acid at a pH of 4-5 through vacuolar proton pumps, which also tilt the bone metastasis environment towards acidity (29). Thus, it is thought that AO accumulates at sites of bone metastases in an acidic environment. However, it was feared that the reduction of osteoclasts by ZOL may reduce the acidic environment of bone metastasis sites and reduce the accumulation of AO, thereby weakening the anti-tumour effect of ZOL. Nevertheless, because the AO+ZOL group also experienced a reduction in local tumour lesions, ZOL administration may not interfere with the effect of AO. Similarly, bone destruction was significantly reduced in the AO+ZOL group, suggesting that AO may not inhibit the effect of ZOL. Therefore, the combination of AO and ZOL may be more effective because of the interaction between the local tumour suppression effect of AO and the bone destruction suppression effect of ZOL.
In patients with carcinomas, multiple bone metastatic lesions can cause pain at various sites and progressively decrease activities of daily living. In addition, the metastasis progresses further as the tumour gains momentum, increasing the chance of mortality (1, 3-5). In cases of multiple metastatic lesions throughout the body, haematogenous administration, in which the drug is distributed throughout the body, is considered a beneficial administration method. Thus, the haematogenous administration of AO may be effective for multiple bone metastatic lesions by taking advantage of its accumulative properties. Kusuzaki et al. reported that haematogenous administration of AO to patients with multiple metastases from late-stage cancer reduced symptoms or tumour size in three of five patients (22). This suggests that AO may be effective for extraosseous metastatic lesions such as those in the lungs. At the same time, they reported that low volume haemodilution of AO in patients with multiple carcinoma metastases is likely to be non-toxic to the patient (22). If AO haemodilution is effective for various metastases from any cancer, multiple distant metastases may be treated with AO haemodilution and radiation to the metastatic site.
In this study, the Rad group did not show tumour suppression or other effects. A single radiation dose of 5 Gy is the optimal radiation concentration to induce AO (23). However, higher doses of single irradiation or fractionated irradiation are used as radiotherapy for breast cancer (30, 31). Studies have also reported that breast cancer becomes radioresistant when it metastasizes (32) and that E0771 cells are resistant to γ-ray therapy up to 30 Gy (33). Therefore, a single radiation dose of 5 Gy in this study did not show any tumour-suppressive effect.
This study is limited in that it only evaluated breast cancer cells. There are various types of carcinomas, such as renal and lung cancer, and the efficacy of these treatment modalities may differ according to the primary tumour. In addition, this study did not conduct histological studies on early bone metastatic lesions. At 6 weeks post-treatment, bone destruction was so severe that it was difficult to assess osteoclasts, which cause progressive bone destruction. In the future, the relationship between AO, ZOL, and osteoclasts, among other factors, should be investigated by using tissue from early-stage bone metastasis and conducting histological evaluation, including acidity.
Conclusion
Haematogenous administration of AO was effective in reducing local bone metastatic lesions. In addition, ZOL showed an inhibitory effect on bone destruction, and the combined use of AO and ZOL showed higher efficacy against bone metastases. In the future, the efficacy of AO and ZOL in various cancers and metastatic sites will be investigated, and this may lead to a new cancer treatment.
Acknowledgements
This research was funded by JSPS Kakenhi Grants (Number 19K09640).
Footnotes
Authors’ Contributions
Conceptualization, Yuji Kasukawa and Naohisa Miyakoshi; Data curation, Ryo Shoji; Investigation, Ryo Shoji, Hikaru Saito, Kazunobu Abe, Shun Igarashi, Shuntaro Harata and Fumihito Kasama; Methodology, Hiroyuki Tsuchie and Hiroyuki Nagasawa; Project administration, Hiroyuki Tsuchie; Resources, Ryo Shoji; Supervision, Naohisa Miyakoshi; Validation, Michio Hongo, Koji Nozaka and Daisuke Kudo; Writing – original draft, Ryo Shoji; Writing – review & editing, Hiroyuki Tsuchie.
Conflicts of Interest
The Authors declare no conflicts of interest.
- Received June 15, 2022.
- Revision received September 10, 2022.
- Accepted September 12, 2022.
- Copyright © 2022 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
