2D and 3D In Vitro Co-culture for Cancer and Bone Cell Interaction Studies

Abstract

Co-culture assays are used to study the mutual interaction between cells in vitro. This chapter describes 2D and 3D co-culture systems used to study cell-cell signaling crosstalk between cancer cells and bone marrow adipocytes, osteoblasts, osteoclasts, and osteocytes. The chapter provides a step-by-step guide to the most used cell culture techniques, functional assays, and gene expression.

1. Introduction

Primary bone cancer (sarcoma), multiple myeloma, and cancers that originate in a distant site and preferentially metastasize to bone (such as breast and prostate cancer) severely disrupt the normal bone microenvironment [1]. Malignant tumor cells that colonize the bone cause bone lesions by physical interaction (juxtacrine signaling) or through the secretion of factors that alter the equilibrium between bone formation and bone resorption (paracrine signaling), resulting in disrupted bone remodeling [2]. As a result of increased resorption by osteoclasts (osteolysis), bone-derived factors released from the bone matrix induce tumor cell proliferation and survival, exacerbating bone turnover and damage, through what is termed the ‘vicious cycle,’ and commonly seen in myeloma and bone metastatic breast cancer [3, 4]. On the other hand, osteoblastic lesions are caused by excessive osteoblastic activity, decreased osteoclast activity, and the accumulation of weak bone matrix, and are more commonly observed than osteolytic lesions in prostate cancer. The bone microenvironment has also been identified as one of the key contributors to cancer resistance to radio and chemotherapy treatments. Moreover, it is well-recognized that the microenvironment plays a crucial role in tumor development, progression, and metastatic spread [5]. In vitro, cell-based models often utilize only one cell type and do not accurately reproduce the complexity of the in vivo environment where multiple cells interact and signal simultaneously. Thus, they often fail to predict in vivo cascades accurately. For this reason, many in vitro culture models that recreate the tumor microenvironment and are less expensive than in vivo (mouse) models have been developed and optimized recently.

Co-culture systems are in vitro cultures in which two or more different types of cells are cultured together in the same dish/well [6]. Co-cultures are divided into direct culture, where two-cell types are layered one on top of the other, or cultured as a mixed population, allowing cell-to-cell interactions; or indirect, where membrane inserts or pre-conditioned media allow for separated culture of the two-cell populations [7–12]. Direct co-cultures are utilized for studies requiring physical interaction between the two-cell populations, e.g., studying adhesion molecules, cytokine production, and juxtacrine signals. Cell-to-cell direct cultures present technical difficulties, and more optimization steps are often needed. Indirect co-cultures are sometimes preferred as they provide more reproducible results, but these do not capture cell-cell adhesion/integrin/notch-type signaling. Indirect co-cultures are achieved by using inserts with a permeable membrane that allows secreted soluble factors to diffuse through the membrane or using previously produced conditioned medium from one of the cell types of the co-culture. Indirect co-cultures are useful for studies evaluating paracrine signaling on population-specific cellular changes, but conditioned media co-cultures ignore the dynamic feedback loop that is likely occurring between the cells.

More recently, three-dimensional (3D) cell-culture techniques have been successfully developed, allowing researchers to better reproduce the spatial and physical conditions in the bone marrow microenvironment [6, 13–15]. However, compared to 2D culture, in which tumor and bone cells are cultured on a flat, tissue culture surface, 3D techniques better recapitulate the microenvironment within the body, offering a more realistic in vitro model system. Thus, 3D culture allows for studying cell function, behavior, morphology, gene expression, and paracrine and cell-cell contact effects in a fashion more representative of in vivo conditions. Compared to mouse models, 3D co-cultures are more affordable and facilitate the study of cell communication between cell types, otherwise masked by the multiple and simultaneous interactions that occur in the tumor microenvironment. This chapter describes general methodological guidelines for setting up and evaluating cell-cell interactions using in vitro 2D and 3D co-culture systems.

2. Materials

All tissue culture procedures are performed under sterile conditions in a laminar flow tissue culture hood. Co-cultures are maintained at 37°C and 5% CO₂. Animal and human samples must be collected and processed per institutional- and national-specific guidelines and regulations.

2.1. General Reagents/Materials/Equipment

1. Incubator and cell-culture flow cabinet.

2. Sterile tissue culture dishes/plates/flasks.

3. Falcon tissue culture plates (see Note 1).

4. Conical polypropylene centrifuge tubes.

5. 0.4 μm pore size transwell chambers.

6. Sterile instruments (scissors, forceps, scalpels, syringes, and needles 19-, 21-, and 25-gauge).

7. Collagen type I (calf skin) coated 10 cm culture dishes and 24-well plates (see Note 2).

8. Lymphoprep mixture or Ficoll-Paque.

2.2. Tissue Culture Reagents

1. Supplemented MEMα medium with 10% fetal bovine serum (FBS), 2 mM L-Glutamine, 100 U/mL penicillin, and 100 μg/mL streptomycin (standard MEM).

2. Supplemented Dulbecco’s modified essential medium with 10% FBS, 2 mM L-Glutamine, 100 UmL/L penicillin, and 100 μgmL/L streptomycin (standard DMEM).

3. Supplemented Roswell Park Memorial Institute (RPMI) 1640 medium with 10% FBS, 100 UmL/L penicillin, and 100 μg mL/L streptomycin (standard RPMI).

4. Supplemented MEMα with 2.5% FBS, 2.5% bovine calf serum (BCS), 100 U/mL penicillin, and 100 μg/mL streptomycin (osteocyte MEM).

5. Phosphate-buffered saline (PBS) pH 7.2.

6. 0.25% Trypsin in PBS.

7. Osteogenic medium: standard MEM supplemented with 10 mM Beta-glycerophosphate disodium salt hydrate (βGP) and 50 μg/mL L-ascorbic acid (optional addition of 10–100 nM dexamethasone; see Table 1).

8. Adipogenic medium: standard MEM supplemented with 10 μg/mL insulin, 0.5 μM isobutylmethylxanthine (IBMX), 0.25 μM dexamethasone, 0.1 μM indomethacin, and Troglitazone (1 μM).

9. Collagenase solution: 1 mg/mL collagenase type IA (Sigma) and sterilized using a 0.22-μm filter. Make fresh before use. Stock 10 mg/mL solution can be prepared in advance, sterile filter, and the aliquots stored at −20 °C.

10. Alizarin Red solution: 40 mM in distilled H₂O, adjust the pH between 4.1 and 4.3 with ammonium hydroxide (10% v/v).

11. Alkaline phosphatase lysis buffer: 0.05% Triton X-100 added to DEA/MgCl2 buffer (1 M diethanolamine and 1 M MgCl₂ made up in 100 mL dH2O and pH adjusted to 9.8).

Table 1.

Working and stock solutions of reagents

Supplements	Working solution	Stock solution and storage
β-glycerophosphate	10 mM	100 mM in MEMα. Aliquots stored at −20 °C.
L-ascorbic acid	50 μg/mL	1 mg/mL in MEMα. Aliquots stored at −20 °C
Dexamethasone	1–100 nM	10 μM in absolute ethanol. Aliquots stored −80 °C
1,25(OH)2D3	10 nM	10 μM in absolute ethanol. Aliquots stored −80 °C
PGE2	1 μM	10 mM in absolute ethanol. Aliquots stored −80 °C
Insulin	1–10 μg/mL	Supplied as 10 mg/mL. Store at 4 °C
Indomethacin	10–50 μM	100 mM in DMSO. Protect from light. Aliquots stored at −70 °C
IBMX	0.25–0.5 mM	250 mM in DMSO. Aliquots stored at −20 °C
Troglitazone	1–10 μM	10 mM in DMSO Aliquot, store at −20 °C
M-CSF	20–50 ng/mL	Reconstitute 50 μg/mL in 0.1% BSA (PBS). Aliquots stored at −80 °C
RANKL	10–120 ng/mL	Reconstitute 100 μg/mL in 0.1% BSA (PBS). Aliquots stored at −80 °C

2.3. Tartrate-Resistant Acid Phosphatase (TRAcP) Staining Reagent

1. Naphtol-AS-BI-phosphate: 10 mg/mL of naphthol-AS-BI-phosphate dissolved in dimethylformamide. Prepare in a glass bottle.

2. Veronal buffer (100 mL): in distilled water, dissolve 1.17 g sodium acetate anhydrous and 2.94 g sodium barbiturate (Veronal). Store at 4 °C, protected from light.

3. Acetate buffer pH 5.2 (100 mL): dissolve 0.82 g sodium acetate anhydrous in distilled water and adjust the pH with an acetic acid solution (0.6 mL glacial acetic acid diluted in 100 mL of distilled water). Stored at 4 °C.

4. Acetate buffer 100 mM sodium tartrate (100 mL): In an acetate buffer with pH 5.2, dissolve 2.3 g of sodium tartrate.

5. Pararosanilin: dissolve 1 g in 20 mL of distilled water and add 5 mL of concentrated hydrochloric acid. Heat at 95 °C for 15 min under agitation in a water bath. Filter paper when the solution is at room temperature (RT). Store at 4 °C, protected from light.

3. Methods

3.1. 2D Cancer—Bone Cells Co-culture

3.1.1. Generation of Cancer Cell-Conditioned Medium (CM) (see Note 3)

1. Seed adherent cancer cells in a six-well plate or 1 × 10⁶ /mL for suspension cells.

2. When cancer cells reach 80% confluence, remove the medium and wash it with PBS.

3. Add 2 mL fresh serum-free medium to the cancer cell monolayer. For suspension cancer cells, plate 1 × 10⁶ /mL suspension cells are directly in a serum-free medium. Culture them for 16–24 h at 37 °C, 5% CO₂.

4. Collect conditioned medium and filter through a 0.45 μm filter.

3.1.2. Cancer—Osteoblast Cells Co-culture

The effects of cancer cells on osteoblast cell differentiation and activity can be investigated by direct or indirect co-culturing breast, prostate, or multiple myeloma cancer cells with rodent or human bone marrow stromal cells (BMSCs) (see Notes 4 and 5). Several techniques for isolation and acquisition of primary osteoblasts have been described previously [16, 17] and are described in (Chapter 1) of this edition. The primary osteoblasts obtained with these methods are isolated from different skeletal locations, such as the long bones and calvaria, and sources (human, rat, and mouse).

1. Isolate primary mouse bone marrow stromal cells (mBMSC) from femurs and tibiae of 4- to 12-week-old by cutting off the epiphyses to expose the bone marrow and flushing out the marrow using a 5 mL syringe with a 25-gauge (G) needle and standard MEM.

2. Obtain a single-cell suspension by passing it through decreasing-sized needles (19–21-25G) and transferring it into a conical 15 mL tube.

3. Centrifuge the cell suspension at 300 × g for 3 min and resuspend the cell pellet in 1 mL of standard MEM.

4. Remove red blood cells by Ficoll density centrifugation or by simply culturing the cells overnight in standard MEM, allowing the BMSCs to adhere to the culture plate.

5. The next day, wash the culture to remove the non-adherent cells, harvest the adherent cells, and set up the co-culture as described below.

6. In a six-well plate, seed 1 × 10⁵ BMSCs cells in 2 mL of standard MEM to allow cell growth (Day 1) (see Note 6).

7. When the well is confluent (Day 4), change the medium with 2 mL of osteoblast osteogenic medium and start the co-culture with 10–20% of conditioned medium from cancer cells prepared as previously described.

8. Refresh 50% of the medium, including conditioned medium, every 48–72 h.

9. By Day 10, collect the cell monolayer for osteoblast differentiation marker detection.

10. By Day 21, mature osteoblasts and abundant deposition of unmineralized matrix can be detected in the culture.

11. By Day 28, terminate the culture and proceed with staining the mature mineralized bone nodules.

12. Calvarial-derived osteoblasts isolated from the calvarial bone of 2-day-old mice by sequential collagenase digestion, as described in [18], can also be utilized for these cultures as a source of osteoblasts (see Fig. 1).

Fig. 1.

Timetable of bone marrow—cancer cell-conditioned medium co-culture. See the text under Subheadings 3.1.2 and 3.1.4 for more details

3.1.3. Osteoblast-Like Cells and Cancer Cells Indirect Co-culture

1. While primary osteoblasts are the preferred cell source to investigate cancer cell effects on osteoblasts, pre-osteoblast MC3T3-E1 sub-clone 4 and the osteoblast-like osteosarcoma cell lines, SaOS2 and MG-63 are also widely used because of their availability, rate of proliferation, expression of osteoblastic marker gene and response to several osteogenic signaling molecules [19–21]. In a six-well plate, seed 1 × 10⁵ in 2 mL of complete medium (MC3T3-E1 sub-clone 4 in standard MEM and SaOS₂ in standard DMEM) until the cells are confluent (Day 1).

2. By Day 3, the monolayer should be 80–90% confluent. Add 1000 adherent cancer cells or suspension cells in a ratio of 1:10. Replace the complete proliferation medium with osteoblast osteogenic medium containing 1% FBS and supplemented with 2 mM β-GP and 50 μg/mL ascorbic and 10 nM dexamethasone in the case of Saos-2 (see Notes 7 and 8).

3. Refresh media every 48–72 h.

4. Between Day 4 and Day 7, collect the cell monolayer for osteoblast differentiation marker detection.

5. By Day 12–14, terminate the culture and proceed with staining the mature mineralized bone nodules.

3.1.4. Bone Marrow-Derived Adipocytes—Cancer Cell CM Co-culture

1. As described above, isolate primary mouse bone marrow stromal cells (mBMSC) from femurs and tibiae.

2. In a six-well plate, seed 1 × 10⁵ BMSCs cells in 2 mL of standard MEM to allow cell growth (Day 1).

3. When the well is 80% confluent (Day 4), change the medium with 2 mL of adipogenic medium and begin the co-culture with 10–20% of conditioned medium from cancer cells prepared as described previously.

4. On Day 6, refresh 50% of the medium with an adipogenic medium supplemented with freshly prepared cancer cell-conditioned medium to the culture (see Notes 9 and 10).

5. At Day 8, refresh 50% of the medium with standard MEM supplemented with 10 μg/mL insulin and Troglitazone 1 μM. Add freshly prepared cancer cell-conditioned medium to the culture.

6. At Day 11, refresh 50% of the medium with standard MEM supplemented with 10 μg/mL insulin.

7. By Day 12, terminate the culture and proceed with the staining of the mature adipocytes (see Fig. 1).

3.1.5. Mouse Pre-adipocyte—Cancer Cell Direct Co-culture

1. In a 12-well plate, seed 1 × 10⁵ 3 T3-L1 cells in 1 mL high glucose DMEM supplemented with 10% FBS, 2 mM L-glutamine, 100 U/mL penicillin, and 100 μg/mL streptomycin (proliferation medium).

2. Allow the cells to grow to 100% confluence in the proliferation medium (Day 1).

3. After 2 days (Day 3) post confluence, replace the medium with proliferation medium supplemented with 10 μg/mL insulin, 0.5 μM IBMX, and 1 μM dexamethasone and add 10–20% of conditioned medium from cancer cells prepared as described before.

4. At Day 5, replace the medium with a proliferation medium supplemented with 10 μg/mL insulin and a freshly prepared cancer cell-conditioned medium. Maintain the adipocyte culture under this condition for another 4 days by replacing 50% of the medium every 48 h.

5. At Day 9, remove the entire medium, replace it with a proliferation medium supplemented only with a cancer cell-conditioned medium, and grow the cells under these conditions until adipocytes are fully differentiated by changing the medium every 48 h.

6. By Day 12–14, terminate the culture and proceed with the staining of the mature adipocytes.

3.1.6. Osteoclast Precursor, Osteoblast, and Cancer Cell Co-cultures

The co-culture of cancer cells, primary calvarial osteoblasts, and osteoclast precursors can be used to investigate how cancer cells influence osteoblasts and stromal cells to support osteoclast formation (see Note 11) (see Fig. 2).

Fig. 2.

Timetable of osteoblast and osteoclast precursors—cancer cell co-culture. See the text under Subheading 3.16 for more details

1. Isolate osteoblasts from calvaria of 2-day-old mice using serial collagenase digestion as described in [18].

2. Maintain osteoblast culture under standard conditions for 3 days until confluence.

3. Collect the osteoblast cells by centrifugation at 300 g × 5 min and estimate the cell number using a hemocytometer.

4. Day 1, seed calvarial osteoblasts into a 96-well plate at 8 × 10³ in 150 μL of culture medium supplemented with 10 nM of 1, 25(OH)₂D₃ and 1 μM PGE₂.

5. Day 2: isolate bone marrow cells as described above.

6. From each well, remove 100 μL of culture medium and replace with bone marrow cell suspension 150–200 × 10³ total bone marrow cells per well in 50 μL of culture medium supplemented with 10 nM of 1,25(OH)₂D₃ and 1 μM PGE₂ and 50 μL of culture medium containing cancer cell-conditioned medium (5–20% v/v) obtained as previously described.

7. Refresh the medium every 48 h supplemented with freshly prepared cancer cell-conditioned medium until multinucleated osteoclasts form—normally around Day 7 or 8 (see Note 12).

3.1.7. Purified Osteoclast Precursor and Cancer Cell Co-cultures

To study the direct effects of cancer cells on osteoclast formation and bone resorption, purified primary osteoclast precursors isolated from the long bones of 4–12-week-old mice or human buffy coat/peripheral blood mononuclear cells (PBMCs) can be utilized. Please refer to Chapter 2 for a detailed description of the isolation of mouse and human osteoclast progenitors, respectively [22, 23] (see Fig. 3).

Fig. 3.

Timetable of purified osteoclast precursors—cancer cell co-culture. See the text under Subheading 3.1.7 for more details

1. Generate the preferred osteoclast precursors (mouse bone marrow or human PMBC) into a petri dish 100 mm and culture in standard MEM in the presence of 25–50 ng/mL of mouse or human M-CSF, according to the source of precursors (see Note 13).

2. After 48–72 h, wash the monolayer with sterile PBS, gently scape the adherent M-CSF-dependent macrophages, and collect them in a 15 mL tube.

3. Spin the cells at 300 g × 3 min and estimate the cell number using a hemocytometer.

4. In a 96-well plate, seed the cells at the following densities in 150 μL of standard MEM containing 20–50 ng/mL of murine or human M-CSF per well:

   1. Mouse M-CSF dependent macrophages: 12 × 10³ cells per well.
   2. Human M-CSF dependent macrophages: 20 × 10³ cells per well (see Note 14).

5. Allow 24 h for cells to adhere.

6. Introduce cancer cells to the co-culture, 250–750 cells per well, depending on the growth of the cell line, by replacing 50% of the culture with a new medium containing cancer cells. If the effects of soluble factors produced by cancer cells on osteoclastogenesis want to be studied, add 5–20% (v/v) of cancer cell condition medium to the culture (see Note 15).

   1. Use ONLY 20–50 ng/mL M-CSF to study the effects of cancer cells on macrophages.
   2. Use 20–50 ng/mL M-CSF and 10–120 ng/mL RANKL to study the effects of cancer cells on osteoclast formation and bone resorption.

7. Refresh the medium every 48 h with a cytokine-containing medium and/or desired cancer cell-conditioned medium.

8. Osteoclasts will begin to form 72 to 96 h following the addition of RANKL (see Note 16).

3.1.8. Bone Marrow Stromal Cells (BMSCs) and Myeloma Cells Direct Co-culture

The binding and adhesion of multiple myeloma cells to bone marrow stromal cells has been shown to activate pathways and the release of cytokines that enhance myeloma cell survival. The use of these co-culture assays can be used to assess the efficacy of novel drugs and identify key molecular players aiding in myeloma survival [24–28].

1. Using bone marrow aspirates from healthy subjects or multiple myeloma patients, perform plasma cell enrichment by magnetic cell separation using anti-CD138 Microbeads according to manufacturer’s instructions.

2. Culture remaining cells and allow adherent BMSCs to expand for 21 days.

3. Plate isolated patient-derived BMSCs or cell line such as HS5 with patient-derived multiple myeloma or myeloma cell lines at a ratio of 4:1.

4. Incubate cells overnight at 37 °C in a humidified 5% CO₂ atmosphere.

5. Treat co-cultures with appropriate treatments and controls and assay viability at chosen time point by microscopy or flow cytometry for viability or expression of chosen markers.

3.1.9. Osteocyte and Myeloma Cells Direct Co-cultures

We describe different methods to study the interactions between osteocytes and myeloma cells (see Note 17). The combined use of the described co-culture strategies allows the identification of the nature of these interactions (cell-to-cell interactions versus exchange of soluble factors, see Fig. 4) and the consequences they have on osteocyte gene expression and viability [29–34].

Fig. 4.

Schematic representation of myeloma-osteocytes co-culture. See the text under Subheading 3.1.9 for more details

1. Day 1, plate osteocytes on 24-well collagen1-coated plates at a 30,000 cell/cm density in complete MEM (see Note 18).

2. Incubate cells overnight at 37 °C in a humidified 5% CO₂ atmosphere to allow attachment of osteocytes.

3. On Day 2, discard the media and gently wash the attached osteocytes with PBS.

4. Add 300 μL of myeloma cell suspension (3×10⁵ myeloma cells suspended in standard MEM) to the osteocytes and incubate at 37 °C in a humidified 5% CO₂ atmosphere to allow cell-to-cell interaction and exchange of soluble factors between osteocytes and myeloma cells (see Note 19).

5. To separate osteocytes from myeloma cells, first remove the medium containing myeloma cells, wash osteocytes gently with PBS once, and then incubate for 5 min at 37 °C with EDTA to remove the remaining myeloma cells (see Fig. 5 and Note 20).

Fig. 5.

Osteocyte apoptosis induced by myeloma cells in co-culture models. See the text under Subheading 4 for more details

3.1.10. Osteocyte and Myeloma Cells Co-cultures Using the Chamber System

1. Day 1. Plate osteocytes on collagen-coated plates at a 30,000 cell/cm density in a complete MEM medium.

2. Incubate cells overnight at 37 °C in a humidified 5% CO₂ atmosphere to allow attachment of osteocytes.

3. Day 2. Discard the media and gently wash the attached osteocytes with PBS. Add 300 μL of complete MEM media to the well.

4. Add 150 μL of myeloma cell suspension (3 × 10⁵ myeloma cells suspended in complete MEM medium) to the transwell insert and place it on top of the osteocytes (see Note 20).

5. Remove the transwell insert to separate osteocytes from myeloma cells and wash the osteocytes with PBS once (see Note 21).

3.2. Characterization of 2D Cancer—Bone Cells Co-culture

3.2.1. Alkaline Phosphatase Activity

This assay is based on converting p-nitrophenol phosphate (colorless) into p-nitrophenol (yellow) by the alkaline phosphatase enzyme, highly expressed by differentiated osteoblasts. To identify the effects of cancer cells on osteoblast differentiation:

1. Once the co-culture is terminated, remove the medium from the wells and wash once with pre-warmed PBS.

2. Resuspend the osteoblast layer in 150 μL of ALP lysis buffer (for a 96-well plate or 500 μL for a 12-well plate) and incubate for 15 min at RT. The lysates can be stored at −20 °C.

3. In a fresh 96-well plate, add 50 μL of osteoblast lysate to the wells (see Note 22).

4. Add 50 μL of substrate solution (20 mM p-nitrophenol-phosphate resuspend in DEA/MgCl₂ buffer pH 9.8) to the wells and immediately measure absorbance at 405 nm at 37 °C at 5 min intervals for 30 min.

5. Alkaline phosphatase activity is normalized to cell number as determined by AlamarBlue assay.

3.2.2. Bone Nodule Formation Assay

This protocol is used to identify the effects of cancer cells on osteoblast activity and the ability of osteoblasts from mouse long bone and calvaria and human bone marrow to form mineralized bone nodules.

1. Once the co-culture is terminated, remove the medium containing the cancer cell-conditioned medium or the insert containing cancer cells and carefully wash the osteoblast monolayer three times with pre-warmed PBS.

2. Add 2 mL/well of 70% pre-chilled ethanol and fix the cells for at least 1 h at 4 °C (see Note 23).

3. Remove ethanol 70% from the wells and wash four times with 2 mL/well dH₂O.

4. Add 1 mL/well of Alizarin Red solution and incubate for 20 min at room temperature on a rocker.

5. Remove the Alizarin Red solution and wash at least four times with 2 mL/well dH₂O—5 min on a rocker for each wash until the excess staining is removed.

6. Leave the plates to dry at an angle overnight before proceeding with the destaining and quantification.

7. Destaining:

   • Method A: Add 1 mL of 10% (w/v) cetylpyridinium chloride to each well in 10 mM sodium phosphate (pH 7.0). Incubate for 15 min on a rocking platform at RTand measure the absorbance of the extracted stain using a spectrophotometer at 562 nm and compare it to an Alizarin Red S standard curve (0–10 mM).
   • Method B: Add 800 μL of 10% acetic acid (v/v) to each well and incubate for 30 min on a rocking platform at RT. Collect supernatant in an Eppendorf and wash the well with an additional 200 μL of 10% acetic acid (v/v). Add 250 μL of mineral oil to the tube (to avoid evaporation). Incubate the tube at 85 °C for 10 min and then place it in ice for an additional 5 min. Spin the tube at 12000 rpm for 15 min. Transfer 250 μL of the supernatant to a fresh tube, add 100 μL of 10% ammonium hydroxide, and mix gently. Using a spectrophotometer, measure the absorbance of 100 μL of the final solution at 405 nm.

3.2.3. Viability/Proliferation of Osteoclast Precursors

The effects of tumor-derived factors on the proliferation of osteoclast precursors can be easily determined through measurement of cell viability/proliferation with commercially available assays such us Alamar Blue, MTT, or Thymidine incorporation assays.

1. Plate osteoclast precursor cells at densities described above.

2. Expose to cancer cell-conditioned medium at varying concentrations of 10–75% (v/v).

3. Measure cell viability/proliferation after 24–96 h.

3.2.4. Tartrate-Resistant Acid Phosphatase (TRAcP) Staining

Osteoclasts express high levels of TRAcP; thus, staining of osteoclasts for TRAcP expression is a quick and easy way to identify multinucleated osteoclasts. TRAcP-positive cells with three or more nuclei are considered osteoclasts. The number and size of osteoclasts and the number of nuclei determine the effects of cancer cells on osteoclastogenesis.

1. To terminate the culture, carefully remove the culture medium, wash with PBS, and fix in 100 μL of 4% formaldehyde at room temperature for 15 min.

2. Rinse wells with PBS twice, and either stain immediately or store at 4 °C overnight.

3. Add 50–100 μL of TRAcP stain solution per well.

4. Incubate at 37 °C for 30–45 min.

5. When the desired staining level is obtained, remove the TRAcP stain solution from the wells and wash it thrice with PBS. Add 200 μL of 70% ethanol and store at 4 °C.

3.2.5. Resorption Pit Assay

Tumor cells within the bone microenvironment not only induce the formation of osteoclasts but can also enhance their activity. Osteoclast activity can be measured using the resorption pit assay. Mature osteoclasts, obtained using the abovementioned method, are lifted from the culture dish using trypsin-EDTA solution. Cells are then plated on dentine or bone slices or in Corning OsteoAssay surface multi-well plates in the presence of cancer cell condition medium and allowed to resorb bone for 48 h. Pits can be analyzed as follows:

1. Dentine slices.

   1. TRAcP stain osteoclasts, then thoroughly clean the slices with tissue paper to remove adherent cells.
   2. Visualize resorption pits using a light-reflected microscope. Resorption pits will appear dark in comparison to unresorbed areas.
   3. Quantify the resorbed areas using ImageJ or image analysis software.
   4. Express the result as the number of resorption pits per osteoclast area.
2. Bone slices:

   1. Vigorously clean the slices with a cotton swab.
   2. Stain the slices with 1% toluidine blue to reveal resorption.
   3. Quantify the pit area using visual image analysis software as above.
3. Osteo Assay surface multi-well plates (see Note 24).

   1. TRAcP stain osteoclasts and quantify them.
   2. Incubate with 50% bleach for 10 min to remove osteoclasts and rinse thrice in PBS.
   3. Fix in 70% (v/v) ethanol.
   4. Resorption pits can be visualized using phase contrast microscopy.

3.2.6. Determination of Changes in Osteocyte Proliferation and Viability

Here, we describe two methods to determine the effects of osteocyte-myeloma cell interactions on proliferation and cell viability in vitro (see Note 25).

1. Changes in Plasma Membrane Permeability.

   1. At the end of the culture period, harvest osteocytes by adding 250 μL of trypsin/well and incubate for 5 min at 37 °C until the osteocytes detach.
   2. Collect the cells by adding 0.25 mL of standard MEM/well.
   3. Spin the cells 300 g × 5 min and resuspend the pellet in 100 μL of PBS.
   4. Mix 10 μL of Trypan Blue with 10 μL of the cell suspension and transfer the mix to a hemocytometer. Count living (clear) and dead (blue) cells and calculate the number of total cells alive (proliferation) and the percentage of dead osteocytes (viability) for each condition (Fig. 5a; see Note 26).
2. Transfection with Nuclear Fluorescent Proteins.

   1. Day 1. Plate 10⁶ cell/cm osteocytes in standard MEM in a collagen-coated 10 cm culture dish and allow them to attach overnight.
   2. Day 2. Mix 20 μL of Plus reagent, 7.8 μg of a plasmid containing nuclear GFP, and 1.3 mL of serum-free MEM. Incubate at room temperature for 15 min.
   3. Add 2.7 mL of serum-free MEM plus 10 μL of Lipofectamine and incubate at room temperature for 15 min.
   4. Remove the medium from the cell monolayer, wash once with serum-free MEM, and transfer the DNA-Plus reagent-Lipofectamine mixture (5 mL). Incubate for 3–6 h (see Note 27).
   5. Remove media and add 10 mL of complete MEM medium. Culture overnight.
   6. Day 3. Check the cells under a fluorescent microscope to determine the efficiency of the transfection (see Note 28).
   7. Harvest the osteocytes and plate them on 24-well collagen-coated plates at a 30,000 cell/cm density in a complete MEM medium.
   8. Establish the co-cultures with myeloma cells as described previously and maintain them in culture for the desired time.
   9. Spin the plate for 5 min at 1200 rpm to stop co-culture and carefully remove the supernatant.
   10. Fix the cells for 8 min with 250 μL/well of 3.7% formaldehyde in PBS. Removing myeloma cells is unnecessary when working with direct co-cultures (see Fig. 5c).
   11. Remove the fixative and wash once with 250 μL/well of PBS. Add 500 μL/well of PBS containing 0.01% thimerosal. Store at 4 °C.
   12. Calculate the percentage of cells exhibiting chromatin condensation and nuclear fragmentation by counting at least 250 cells from random fields for each experimental condition.

3.3. 3D Cancer—Bone Cells Co-culture

Silk protein has long been used in biomedical research due to its useful properties. It is a strong, biocompatible, relatively inexpensive natural polymer that can be formulated into many shapes and types of materials, ranging from silk scaffolds [35, 36] to hydrogels [37], electro-spun mats [38] to tubes [39], nanoparticles for drug delivery [40] to 3D printed shapes and molded screws for use as steel alternatives in orthopedic surgery [41]. Silk tissue engineering has been useful in skin [42], adipose [43], vascular, cartilage [44], and osteochondral [45] tissue engineering. Over the past 20 years, silk scaffolds have been especially useful in tissue-engineered bone applications [46–48]. A wide array of types of silk scaffolds have been shown to support osteogenic differentiation as shown through cell expression of bone markers (e.g., osteopontin, bone sialoprotein, and collagen I), mineralization (micro CT, calcium deposition, and von Kossa and Alizarin Red staining) mechanical properties, and bone enzymatic activity (alkaline phosphatase activity) [48, 49]. Silk scaffolds have been used for models of breast [50] or prostate [51] cancer-bone metastasis, as well as multiple myeloma [48]. Here, we provide protocols for making silk scaffolds and using these to make tissue engineered models of bone and cancer (Fig. 6).

Fig. 6.

Schematic representation of silk scaffolds for 3D tissue-engineered bone and cancer co-culture models. See text under Subheading 3.3 for more details

3.3.1. Materials and Suggested Sources

1. Raw Silk Cocoons: These can be obtained from many sources.

2. For Boiling Cocoons:

   1. Griffin Low Form Beaker-2 L.
   2. Sodium Carbonate.
   3. Spinner/Hot Plates.
   4. Tinfoil, heavy-duty.
3. For Lithium Bromide Dissolving and Dialysis:

   1. Lithium Bromide ReagentPlus.
4. Buoys For Boiling Cocoons:

   1. Griffin Low Form.
5. For Dialysis using Cassettes:

   1. Single-Use Needles, BD Medical.
   2. Slide-A-Lyzer Dialysis Cassettes, 3.5 K MWCO.
   3. Syringe with Luer-Lok^® Tip, 3 mL and 20.
6. For Dialysis using Tubing:

   1. Spectra/Por^® 3 Dialysis Membranes, MWCO 3500.
   2. Spectra/Por^® Universal Closures.
7. For Forming Silk Scaffolds:

   1. NaCl.
   2. Molds: VIAL SNAP CAP LDPE.
   3. Salt Sieves (600 microns) or (500 microns).

8. Weight boxes (small, medium, and large), scissors, centrifuge tubes, magnetic stir bars, flat/spoon spatulas (stainless steel), 150 mL glass beaker, hot hand protectors, parafilm, petri dishes (also can be used as a mold for silk scaffolds), rubber bands and ruler.

3.3.2. Silk Scaffold Fabrication and Seeding for Tissue-Engineered-Bone

Aqueous (Water-Based)-and HFIP (Hexafluoroisopropanol)-Derived Silk Processing Steps

1. Obtain Silk Cocoons. Silk fibroin scaffolds are fabricated using silkworm (Bombyx mori) cocoons (see Fig. 1).

2. Boil Cocoons. Five grams of silkworm cocoons, with the silkworm removed, are cut into small pieces, placed into 2 l of boiling 0.02 M Na₂CO₃, and boiled for 30 min. Discard the supernatant containing silk sericin, retain and wring out silk fibroin, and rinse this in fresh deionized water for 20 min three times. Wring out the silk and lay it out to dry overnight.

3. Dissolve the dry silk fibroin. In a glass beaker, mix a 9.3 M solution of LiBr to dried silk with a 20% weight silk/volume LiBr ratio. Cover this and place it at 60 °C for 4 h or until dissolved.

4. Perform Dialysis to remove ions from the silk solution. Fill dialysis cassettes or dialysis tubing (3500 MW cutoff) with dissolved silk/LiBr solution using the volume of solution recommended on the container. Dialyze 12 mL of silk LiBr solution against 1 L deionized water for 3 days, changing the water ~6 times.

5. Purify silk. Remove the silk solution from the dialysis cassettes, centrifuge for 20 min (9000 rpm, 5–10 °C) twice, and pour off the supernatant to remove debris. Filtering is also possible.

6. Test the concentration of silk using either wt/wt% or wt/vol% measurements. To do this, weigh 1 mL sample of the solution before and after drying for 12 h at 60 °C, divide the dry weight by the wet weight, and multiply by 100%. As another option, dry weight (g) per 1 mL, times 100%, provides wt/vol% concentrations. Concentrate silk by allowing for evaporation or dilute it by adding water to get to the desired concentration.

Aqueous-Derived Silk Fibroin Scaffold Preparation

1. Sift NaCl particles using sieves to capture crystals of the desired pore size (often ~500–600 μm diameter).

2. Add 2 mL of silk solution to molds and slowly sprinkle 4 g of sifted NaCl particles into the silk solution. Silk particles should fall through the silk solution as single particles and be spread as evenly as possible. The silk and salt ratio should be constant, but the total amounts can be increased when using larger molds or petri dishes (see Note 29).

3. Snap the molds closed or seal molds so they are airtight and leave at room temperature for 24–48 h.

4. Once hardened, put scaffolds in deionized water for 2 days to extract NaCl.

5. Remove scaffolds from containers and rinse in water for 1 more day to be sure all salt is removed.

6. Cut scaffolds to shape, being sure to cut all edges so that cells can access the porous network inside the scaffolds.

HFIP-Derived Silk Fibroin Scaffold Preparation

1. Lyophilize silk fibroin aqueous solution by pouring it into a 50 mL conical tube (do not fill), covering it with a Kimwipe secured with a rubber band, freezing it at −80 °C overnight or until fully dried, and placing it in a lyophilizer.

2. Once silk appears completely lyophilized, re-dissolve silk with hexafluoroisopropanol (HFIP) to produce a 17 wt/wt% HFIP-derived silk solution in a chemical hood (see Note 30).

3. Sift NaCl particles using sieves to capture crystals of the desired pore size (often ~500–600 μm diameter).

4. Pour HFIP-silk solution (2 mL) over 4 g of NaCl sifted particles in small plastic containers. Use a 3 mL plastic syringe for this step, moving quickly as HFIP will evaporate fast. Cap all containers when not being used. HFIP-silk solution will be very sticky, so be cautious.

5. Cap the containers quickly and leave them overnight at room temperature.

6. Open containers to allow HFIP to evaporate from scaffolds for 3 days.

7. Submerge scaffolds in a 90% methanol solution for 30 min to 3 days to induce β-sheet structure formation.

8. Remove samples from methanol and immerse them in deionized water for 3 days using a stir bar in a large beaker, changing water twice daily to remove NaCl particles.

9. Remove scaffolds from containers and rinse them in water for 1 more day to be sure all salt is removed.

10. Cut scaffolds to shape, being sure to cut all edges so that cells can access the porous network inside the scaffolds.

Tissue-Engineered Bone from Bone Marrow-Derived MSCs

1. Obtain scaffolds of equal sizes using a biopsy punch or razor blade to create ~4 mm diameter × 4 mm height for in vitro applications or any size desired.

2. Autoclave or soak in ethanol to sterilize before use with cells.

3. To sterilize the scaffolds (if dry), scaffolds can be wrapped in aluminum foil and placed into autoclavable packets or, if wet, put into a small glass, water-filled jar and autoclaved.

4. Soak sterile scaffolds in media containing FBS for 24 h before seeding to allow proteins from the FBS to adhere to the scaffolds.

5. Seed cells to scaffolds using high cell concentrations and small volumes, so the cells are encouraged to adhere to the scaffold in sterile cell-culture well plates in a biosafety cabinet (see Notes 31–33).

6. Place scaffolds seeded with cells into the incubator (37 °C, 5% CO₂) for 1–2 h to allow cells to adhere to scaffolds.

7. Add fresh media to wells to cover the scaffolds.

8. After cells spread throughout the scaffold, add osteogenic medium (if differentiation is needed) to the scaffolds. Cells are differentiated for as long as desired to create tissue-engineered bone (TE-bone).

9. Add tumor cells and image scaffolds using live-dead imaging on a confocal microscope to determine if tumor cells are growing well with the bone cells (see Note 34).

10. Change the medium every 3 days or as often as needed.

11. Quantify changes in mineralization using μCT, histology, or other bone protocols (see Note 35).

12. Quantify changes in tumor cells using confocal imaging, bioluminescence imaging, or flow cytometry.

4. Notes

1. Primary bone cells such as calvarial osteoblasts, bone marrow cells, and primary human peripheral blood mononuclear cells (PBMCs) seem to adhere better on Falcon tissue culture plates, leading to better co-culture.

2. Prepare collagen-coated plates as follows: cover the surface of the plates with 0.01% collagen type I solution (in PBS containing 1% glacial acetic acid) and incubate for at least 3 h at 37 °C. Let the plates dry in the tissue culture hood with the UV light on for 30 min.

3. Over recent years, there has been much interest in the role of tumor-derived exosomes. Exosomes can be treated similarly to the conditioned medium and added at the same time points. Methods to isolate exosomes are reviewed here [52].

4. Human stromal cells isolated from the bone marrow aspirates of patients after written informed consent are often used to study the persistent suppression of bone marrow stromal cell differentiation and new osteoblastic bone formation induced by multiple myeloma.

5. Orriss et al. have shown that 10 nM dexamethasone reduces the differentiation of mouse osteoblasts [53]. We recommend adding dexamethasone only in cultures of osteoblast-like cells SaOS2.

6. The alkaline phosphatase (ALP) assay can be used to determine the effects of cancer cells on osteoblast differentiation. In a 96-well plate, plate 5–8 × 10³ cells per well in 150 μL of standard αMEM for 48 h. Once cells are 60–70% confluent, add 5–20% cancer cells conditioned medium and terminate the culture at different times, e.g., 24, 48, 72 h, 7 days.

7. The optimal amount of cancer cells varies per cell line and requires optimization.

8. Co-culture of multiple myeloma and MC3T3-E1 sub-clone 4 cells can be performed utilizing direct co-culture for 1–5 days for differentiation studies (RNA/protein analysis) or by using inserts/transwell (pore size, 0.4 μm) for 12–14 days for bone mineralization studies.

9. Adding Troglitazone or Rosiglitazone 1 μM speeds up the differentiation of BMSCs into adipocytes.

10. Culture media can be fully replaced; however, it is advisable to replace only 50% of the media every time to avoid disrupting/peeling the delicate monolayer.

11. In the absence of primary osteoblasts or stromal cells, it is possible to use the stromal cell lines ST2 and MC3T3-E1 or the osteoblast-like cell lines SaOS2 and MG63.

12. The osteoblast layer is fragile and can easily detach; the utmost care should be taken when changing the medium and stopping the culture.

13. Mouse bone marrow cells or human PBMCs can be directly treated with M-CSF (human or mouse 25–50 ng/mL) and RANKL (human or mouse 10–120 ng/mL) to generate osteoclasts. However, the number of precursors present in these samples is highly variable, and other cells, such as stromal cells and lymphocytes, can affect osteoclast formation. Therefore, we advise normalizing the number of precursors and expanding the macrophage population with M-CSF before adding RANKL. If total mouse bone marrow or human PMBCs are to be used for osteoclastogenesis, we advise plating 150–250 × 10³ cells per well in a 96-well plate. Moreover, when co-culturing with cancer cells, adding M-CSF and RANKL to total PBMC cultures may be omitted. Faust and colleagues showed that culturing PBMCs at high density (~500 × 10³ cells per well in a 96-well plate) generates TRAcP+ osteoclasts capable of bone resorption without the addition of exogenous M-CSF and RANKL [54].

14. CD14⁺ monocyte osteoclast precursors, purified by magnetic sorting from PMBCs, are often utilized for human osteoclast studies. If CD14⁺ monocytes are used for osteoclastogenesis, we advise plating 45 × 10³ cells/well in a 96-well plate.

15. The number of cells per well will depend both on the cancer cell line in question and the duration of the culture; thus, the cancer cell seeding density should be optimized. Examples of cell numbers used: MDA-MB-231—300 cells/well; MCF7—500 cells/well; KHOS 100 cells/well; PC3—300 cells/well; 4 T—200 cells/well; U266—1000 cells/well.

16. An alternative source of osteoclast precursor cells can be found in the mouse and human macrophage/monocyte cell lines, RAW 264.7 and THP-1, respectively. The generation of multinucleated osteoclast-like cells from RAW 264.7 requires only the addition of 10–120 ng/mL RANKL, while no additional M-CSF needs to be added as these cells express both M-CSF and its receptor c-fms. THP-1 cells are a human monocytic leukemia cell line. THP-1 cells can become adherent macrophage-like cells by adding 100 ng/mL of phorbol myristate acetate (PMA) for 24 h. After 24 h, media containing PMA can be removed and replaced with serum-free media for 48 h [55].

17. The methods and cell densities described in this section have been tested for co-cultures between osteocytes (murine primary osteocytes and osteocyte-like cells lines) and myeloma cells (murine and human cells lines and primary CD138⁺ cells from patients), but it can be adapted for other non-adherent cancer cells by varying the ratio between osteocytes and cancer cells and co-culture incubation time.

18. We recommend using a volume of 300 μL/well to plate the cells in 24 well plates. It is important that the osteocytes are equally distributed in the plate and approximately 80% confluent before starting the co-cultures.

19. To establish 24–48 h co-cultures, we recommend using osteocyte MEM for osteocyte-like cell lines and standard MEM for primary osteocytes. We have observed similar results using standard RPMI or a mix of standard MEM and RPMI (50% each). The doubling population of myeloma cells and osteocytes differs, affecting the ratio of osteocyte/myeloma in the co-culture over time. Thus, it is important to adjust the specific osteocyte/myeloma ratio depending on the endpoints studied. The suggested osteocyte/myeloma ratio (1:5) was designed to study osteocyte viability and changes in gene expression in osteocytes. It is also recommended to test different time points since rapid and slow changes in cytokine and signaling pathways occur in these co-culture systems.

20. Controls (osteocytes cultured alone) also need to be washed with PBS and EDTA. We have observed differences in the attachment between osteocytes and myeloma cells when using different myeloma cell lines and primary cells. If needed, repeat the EDTA wash until complete removal of myeloma cells. The use of GFP⁺ cancer cells can be beneficial to assure complete removal, separate cell types by FACS sorting, and determination of cancer cell proliferation when working with mixed populations. Repeated washes with EDTA may cause detachment of osteocytes. If osteocytes detach and/or cancer cells cannot be separated from the osteocytes, we recommend the use of an insert to determine osteocyte viability and the use of species-specific primers to examine gene expression.

21. If working with other cancer cells, it is recommended to determine the ratio of osteocyte/cancer cells and the co-culture time of incubation. Alternatively, conditioned media (CM) collected from cancer cells can be used to determine the effects of cancer-derived soluble factors on osteocyte biology. To collect CM, grow myeloma cells in standard RPMI for 24–48 h. Spin the myeloma cells for 5 min at 1200 rpm and collect CM carefully. CM can be kept at −20 °C until use. Follow steps 1–3 described in Subheading 3.1 to plate osteocytes. Then, add standard MEM mixed with CM (50%) to osteocytes.

22. The study of osteocyte biology is challenging because of their difficult accessibility and the scarce models available for in vitro studies. Several cell lines have been used to characterize the biology of osteocytes: human HOB-01-C1 cells [56], murine MLO-Y4 and MLO-A5 osteocyte-like cells, and murine IDG-SW3 and Ocy454 (these cells can be expanded at 33°C and differentiate into osteocytes when cultured at 37°C under osteogenic conditions). In addition, several methods have been used to obtain and culture primary osteocytes from human and murine bones, providing an alternative to osteocyte-like cell lines [57].

23. In some cases, especially with Saos-2 cells, the cell lysates are too concentrated to use 50 μL of lysate and require dilution in lysis buffer before the assay can be performed.

24. Longer fixation than 1 h is advised, but do not leave cells in the fixation stage for more than 72 h.

25. Osteo Assay surface multi-well plates are coated with an inorganic osteomimetic surface, allowing osteoclast generation and resorption.

26. These methods can also be used to determine the effects of interactions with osteocytes on myeloma cells [58]. To harvest myeloma cells from cell-to-cell contact co-cultures, collect all media containing the myeloma cells and all the washes in tubes previously labelled (put cells from each well in a separate tube). To harvest myeloma cells from chamber systems, collect the media containing the cells from the transwell insert, wash the insert once with PBS, and transfer it to the same tube (put cells from each insert in a separate tube). Spin the myeloma cells for 5 min at 1200 rpm and remove the supernatant carefully. Determine myeloma proliferation/viability and gene expression.

27. We recommend combining different approaches when determining the effects of cancer cells on osteocyte proliferation/viability. Trypan blue determination can underestimate the number of dead osteocytes since most are floating and removed when washing out myeloma cells. Using caspase three inhibitors in the co-cultures can help determine whether decreases in viable cells are due to apoptosis [58].

28. We recommend monitoring the cells hourly during the transfection process. If the cells appear stressed, stop the transfection immediately. The transfection efficiency of the described protocol is about 40% for osteocytes. Note that the percentage of GFP⁺ cells will decrease over time; thus, this protocol is recommended for 24–48 h co-cultures.

29. The use of osteocyte and cancer cells from different species (human versus murine) can be advantageous to determine changes in gene expression if the separation of osteocytes from cancer cells is problematic. If so, we recommend isolating total RNA from the mixed population and analyzing gene expression using species-specific primer and probe sequences.

30. The choice of which type of silk scaffold to use for bone tissue engineering is important for cancer and disease modeling, as well as for tissue regeneration and reconstruction purposes. HFIP-scaffolds and water-based scaffolds have different properties: the mechanical strength of these scaffolds varies, water-based scaffolds break down much faster in the body, and there are many differences in cellular responses (e.g., protein expression, glucose consumption rate, rate of mineralization, etc.) on these different scaffolds [59]. Moreover, other types of silk scaffolds can be developed and haven’t been fully explored in bone-cancer modeling, such as the aqueous-based lamellar scaffolds produced by lyophilizing a frozen aqueous silk solution directly inside a silicon tube cast [49]. Silk pore sizes and solutions can also be modified, although only certain ranges can induce beta-sheets and scaffold formation from silk solution. Also, this can be included in the bone, like silk fibers or scaffolds, which can be functionalized or premineralized [60]. Scaffolds can also be cultured in vitro in bioreactors, spinner flasks, or rocking stages to add fluid flow to the system, encourage osteogenic differentiation, and create an even more realistic environment [61].

31. Silk solutions of 6–8% wt%/wt% work well to construct 500–600 μm pore scaffolds. For a full characterization of which combinations of scaffold pore size and silk concentrations work in HFIP- and aqueous systems [62].

32. All HFIP work should be done in a chemical fume hood.

33. For example, human mesenchymal stem cells (hMSCs) should be seeded at ~17,000 cells/mm³ in 20–40 μL media. Other human or mouse-derived sources (primary cells or cell lines) of osteoprogenitors, mature osteoblasts, or osteocytes can also be explored.

34. Cells that stably express a fluorescent protein, such as GFP or RFP (green or red fluorescent protein), should be used whenever possible. Cell types and concentrations can be modified.

35. The number and type of tumor cells used should be determined based on the hypotheses of interest. If tumor cells are cultured with TE-bone during or after the differentiation process, dexamethasone should not be used in the media, and an acceptable alternative medium for both cell types should be determined.

36. For example, for hMSCs, 1 week of proliferation and 3–4 weeks of osteogenic differentiation produces mineralized, osteoblast-containing scaffolds, depending on the cell donor.