The ex vivo culture of bone explants can be a valuable tool for the study of bone physiology and the potential evaluation of drugs in bone remodeling and bone diseases. The presented protocol describes the preparation and culture of calvarias isolated from newborn mice skulls, as well as its applications.
Bone is a connective tissue constituted of osteoblasts, osteocytes, and osteoclasts and a mineralized extracellular matrix, which gives it its strength and flexibility and allows it to fulfill its functions. Bone is continuously exposed to a variety of stimuli, which in pathological conditions can deregulate bone remodeling. To study bone biology and diseases and evaluate potential therapeutic agents, it has been necessary to develop in vitro and in vivo models.
This manuscript describes the dissection process and culture conditions of calvarias isolated from neonatal mice to study bone formation and the bone tumor microenvironment. In contrast to in vitro and in vivo models, this ex vivo model allows preservation of the three-dimensional environment of the tissue as well as the cellular diversity of the bone while culturing under defined conditions to simulate the desired microenvironment. Therefore, it is possible to investigate bone remodeling and its mechanisms, as well as the interactions with other cell types, such as the interactions between cancer cells and bone.
The assays reported here use calvarias from 5-7 day old BALB/C mice. The hemi-calvarias obtained are cultured in the presence of insulin, breast cancer cells (MDA-MB-231), or conditioned medium from breast cancer cell cultures. After analysis, it was established that insulin induced new bone formation, while cancer cells and their conditioned medium induced bone resorption. The calvarial model has been successfully used in basic and applied research to study bone development and cancer-induced bone diseases. Overall, it is an excellent option for an easy, informative, and low-cost assay.
Bone is a dynamic connective tissue that has several functions, including supporting the muscles, protecting the internal organs and bone marrow, and storing and releasing calcium and growth factors1,2. To maintain its integrity and proper function, bone tissue is continuously under the process of remodeling. In general terms, a cycle of bone remodeling can be divided into bone resorption and bone formation1. An imbalance between these two phases of bone remodeling can lead to the development of bone pathologies. Also, diseases such as breast cancer often affect bone integrity; approximately more than 70% of patients in advanced stages have or will have bone metastases. When breast cancer cells enter the bones, they affect bone metabolism, resulting in excessive resorption (osteoclastic lesions) and/or formation (osteoblastic lesions)3.
To understand the biology of bone diseases and develop new treatments, it is necessary to understand the mechanisms involved in bone remodeling. In cancer research, it is essential to investigate the bone metastasis process and its relation to the metastatic microenvironment. In 1889, Stephen Paget hypothesized that metastases occur when there is compatibility between the tumor cells and the target tissue, and suggested that the metastatic site depends on the affinity of the tumor for the microenvironment4. In 1997, Mundy and Guise introduced the concept of the "vicious cycle of bone metastases" to explain how tumor cells modify the bone microenvironment to achieve their survival and growth, and how the bone microenvironment promotes their growth by providing calcium and growth factors5,6,7.
To characterize the mechanisms involved in bone remodeling and bone metastasis and to evaluate molecules with possible therapeutic potential, it has been necessary to develop in vitro and in vivo models. However, these models currently present many limitations, such as the simplified representation of the bone microenvironment, and their cost8,9. The culture of bone explants ex vivo has the advantage of maintaining the three-dimensional organization as well as the diversity of bone cells. In addition, experimental conditions can be controlled. The explant models include the culture of metatarsal bones, femoral heads, calvarias, and mandibular or trabecular cores10. The advantages of the ex vivo models have been demonstrated in diverse studies. In 2009, Nordstrand and collaborators reported the establishment of a coculture model based on the interactions between bone and prostate cancer cells11. Also, in 2012, Curtin and collaborators reported the development of a three-dimensional model using ex vivo cocultures12. The purpose of such ex vivo models is to recreate the conditions of the bone microenvironment as accurately as possible to be able to characterize the mechanisms involved in normal or pathological bone remodeling and evaluate the efficacy of new therapeutic agents.
The present protocol is based on the procedures published by Garrett13 and Mohammad et al.14. Neonatal mouse calvaria cultures have been used as an experimental model, as they retain the three-dimensional architecture of the bone under development and bone cells, including cells at all stages of differentiation (i.e., osteoblasts, osteoclasts, osteocytes, stromal cells) that lead to mature osteoclasts and osteoblasts, as well as the mineralized matrix14. The ex vivo model does not represent the pathological process of bone diseases totally. However, effects on bone remodeling or cancer-induced bone osteolysis can be accurately measured.
Briefly, this protocol consists of the following steps: the dissection of calvarias from 5-7 day old mice, calvaria preculture, calvaria culture applications (e.g., culture in the presence of insulin, cancer cells or conditioned medium, and even agents with therapeutic potential, according to the aim of the investigation), bone fixation and calvaria decalcification, tissue processing, histological analysis, and result interpretation.
All mice used in these assays were obtained from BALB/c mice strains, using male and female mice indiscriminately. Previous culture experiments have also been performed using other strains, such as FVB, Swiss mice, CD-1, and CsA mice11,12,14. All mice were housed according to National Institutes of Health (NIH) guidelines, Appendix Q. Procedures involving animal subjects have been approved by the Institutional Animals Care and Use Committee (IACUC) at the Center for Scientific Research and Higher Education at Ensenada (CICESE).
1. Calvarial dissection
2. Calvaria culture
3. Culture with cancer cells
4. Fixation
5. Decalcification
6. Tissue processing
7. Embedding
8. Sectioning
9. Staining
10. Quantitative assessment: defining area for analysis
11. Data analysis
To evaluate bone formation in the calvarial model, we cultivated the hemi-calvarias in media with or without 50 µg/mL of insulin. Tissue sections were prepared and stained with H&E. In these conditions, the histology showed that the structural integrity of the calvarial bone was maintained, allowing the identification of its different components (Figure 1). The calvarias treated with insulin presented an increase in the amount of bone tissue compared to the control (Figure 2A). Quantitative histomorphometric analysis for the thickness and bone area of hemi-calvaria confirmed a significant increase for both parameters in the insulin-treated tissues compared to the control (Figure 2B). These data indicate the feasibility of the ex vivo model protocol to reproduce bone formation conditions.
Additionally, the calvarial ex vivo model can be used to reproduce cancer and bone microenvironment conditions. The protocol was used to evaluate the effect of the breast cancer cells MDA-MB-231 on murine calvarias. To perform this, the calvarial bones were directly cocultured with the cancer cells or cultured only with the conditioned media of the cancer cells. After 7 days of culture, the calvarias were embedded in paraffin, and H&E staining was performed. The coculture with MDA-MB-231 breast cancer cells appeared to increase osteolysis, as shown by the decrease in bone and the damages to the calvarial structure when compared to the control calvarias cultivated only with media (Figure 2A). Bone area quantification of calvarias showed a significant decrease in the total bone area of the calvarias cultured with MDA-MB-231 cancer cells compared to the control (Figure 2B). These results demonstrated that the cocultures of cancer cells with calvarial tissue can recreate the cancer and bone microenvironment and could be used to investigate the mechanisms of osteolytic bone metastasis or to evaluate possible inhibitors of this process.
Figure 1: Histological structure of the calvarial bone. Representative tissue section of hemi-calvaria after H&E staining. Please click here to view a larger version of this figure.
Figure 2: Insulin increased the bone area and thickness of hemi-calvarias ex vivo while coculture of hemi-calvarias with MDA-MB-231 breast cancer cells induced osteolysis. For the bone formation model, the hemi-calvarias were cultured in the presence or absence of insulin (50 µg/mL), while for the coculture with cancer cells, hemi-calvarias were cultured in the presence or absence of 5 x 105 MDA-MB-231 cells for 7 days and histomorphometrical analysis was performed on H&E stained sections. (A) Representative tissue sections. (B) Histomorphometrical analysis of the bone area. Results are represented as the average ± SEM (n = 3). *P < 0.05, nonparametric Mann-Whitney U test. Please click here to view a larger version of this figure.
Here, we describe the protocol for a calvarial ex vivo model to evaluate bone formation or resorption and to study the interactions of cancer cells with calvarial mouse bone. The critical steps of this technique are the dissection, culture, embedding, and histomorphometrical analysis of the calvarias. During the dissection of the calvarias, it is crucial to cut the hemi-calvarias into a trapezoid, as it will strongly facilitate the orientation during the paraffin inclusion. When studying cancer cell interactions with the calvarias, it is important to use multiwell plates that are not treated for cell culture to prevent cancer cells from adhering and growing on the plastic instead of the bone. During the tissue inclusion, it is important to carefully orient each hemi-calvaria. They need to be placed in a vertical manner with the sagittal border on the outer side of the paraffin block. The orientation of the bone is essential to obtain consistent histological sections and reproducible results. Similarly, it is critical for consistency during the histomorphometric analysis to define a specific region in the calvaria to measure bone remodeling. Here, we identified first the coronal suture and then the long bone surface where the pictures were taken and the measurements made.
To achieve the standardization and develop the protocols described, we modified some established methodologies slightly. During the dissection of the hemi-calvarias, PBS was used to rinse the mouse heads instead of culture media. To determine the number of cells necessary to produce an effect on the calvarias, the tissue was incubated with different numbers of breast cancer cells. In general, 5 x 105 cancer cells work well for a 7 day assay. Also, during fixation, tissue paper was used instead of sponges to protect the calvarias. The tissue was well-preserved within the paper.
The model described has some limitations. The calvarias from neonatal mice lack some of the components of the bones that are the recipient of metastases, whether structural (i.e., trabecular bone) or cellular components such as immune cells or other cells of the bone marrow, including hematopoietic stem cells that cancer cells interact with. The culture of the calvarias in vitro cannot simulate the complete physiopathology. Also, when it comes to bone metastases, breast cancer cells seldom metastasize to the calvaria, and tend to favor bones with more active bone remodeling, such as the vertebrae, the hip, or the long bones of the arms and legs. Besides, the coculture of hemi-calvarias with cancer cells only represents the latest steps of the metastatic cascade: the colonization of the bone and the growth of the metastasis. Furthermore, the number of samples is limited by the number of pups per litter. It is recommended to use one litter per experiment and not mix litters to avoid variations within the results. We obtained similar results in bone remodeling when using calvarias from BALB/c or FVB mice, but whether the strain affects the time response of bone remodeling in this assay remains to be determined.
The main advantages of the calvarial ex vivo model compared to in vivo models include lower cost, simplicity of the assay, and shorter experimental time to obtain a bone remodeling response. The coculture of cancer cells and calvarias allows studying the cell-contact dependent interactions as well as the effect of secreted factors on bone remodeling, while the use of conditioned media allows focusing on the effect of soluble factors. In addition, the direct contact model can facilitate the characterization of intercellular interactions and molecular mechanisms of the bone metastasis microenvironment, as well as the study and evaluation of new drugs for the treatment of skeletal complications of malignancies.
The authors have nothing to disclose.
The authors thank Mario Nomura, M.D. and Rodolfo Díaz for their help with the histology, and Pierrick Fournier, Ph.D. for his valuable comments to improve the quality of the paper.
24 well cell culture | Corning | CLS3524 | |
24 well non tissue culture | Falcon | 15705-060 | |
2 mL cryovial | SSI | 2341-S0S | |
Antibiotics-Antimycotic | Corning | 30-004-CI | |
BSA | Biowest | P6154-100GR | |
Centrifugue | Eppendorf | 22628188 | Centrifuge 5810R |
Coverslips | Corning | 2935-24X50 | |
Cytoseal resin | Richard Allen | 8310-10 | |
DMSO | D2650-100ML | ||
Dulbecco's Modification of Eagles Medium, with 4.5 g/L glucose and L-glutamine, without sodium pyruvate | Corning | 10-017-CV | |
Dulbecco's PBS (10X) | Corning | 20-031-CV | |
Ebedding Cassettes | Sigma | Z672122-500EA | |
EDTA | Golden | 26400 | |
Embedding Workstation | Thermo Scientific | A81000001 | |
Eosin | Golden | 60600 | |
Ethanol absolute | JALMEK | E5325-17P | |
Fetal Bovine Serum | Biowest | BIO-S1650-500 | |
Filters | Corning | CLS431229 | |
Forceps and scissors | LANCETA HG | 74165 | |
Formalin buffered 10% | Sigma | HT501320 | |
Glass slides 25 x 75 mm | Premiere | 9105 | |
Harris's Hematoxylin | Jalmek | SH025-13 | |
High profile blades | Thermo Scientific | 1001259 | |
Histoquinet | Thermo Scientific | 813150 | STP 120 |
Insulin from bovine pancreas | Sigma | 16634 | |
Microscope | ZEISS | Axio Scope.A1 | |
Microtome | Thermo Scientific | 905200 | MICROM HM 355S |
Mouse food, 18% prot, 2018S | Harlan | T.2018S.15 | |
Neubauer | VWR | 631-0696 | |
Orange G | Biobasic | OB0674-25G | |
Paraffin | Paraplast | 39601006 | |
Paraffin Section Flotation Bath | Electrothermal | MH8517X1 | |
Petri dish | Corning | CLS430167 | |
Phloxin B | Probiotek | 166-02072 | |
Trypan Blue | Sigma | T8154 | |
Trypsin-EDTA | Corning | 25-051-CI | |
Wax dispenser | Electrothermal | MH8523BX1 | |
Xylene | Golden | 534056-500ML |