A 3D Bone Culture Platform Using Human Osteocytes and Decellularized Extracellular Matrix for Modeling Musculoskeletal Diseases

Summary

This protocol describes the fabrication of 3D bone constructs using human osteocytes embedded in methacrylated gelatin (GelMA) hydrogels supplemented with decellularized extracellular matrix. The method is low-cost, scalable, and compatible with basic laboratory equipment, providing a translational platform for studying musculoskeletal disease processes.

Abstract

Modeling musculoskeletal diseases such as osteoporosis requires in vitro platforms that accurately reproduce key features of human bone biology. Conventional 3D culture systems provide valuable insight into cell-cell and cell-matrix interactions, but their dependence on costly materials, bioprinters, or specialized infrastructure limits their broader adoption. Here, we describe a robust and cost-effective protocol for generating 3D bone constructs using human-derived osteocytes embedded in GelMA hydrogels, supplemented with decellularized extracellular matrix (dECM) obtained from donated femoral heads. This pipetting-based method enables the fabrication of reproducible, bone-like constructs at physiological temperature (37 °C), using standard laboratory equipment.

The resulting constructs maintain high viability and structural integrity, supporting downstream applications such as immunofluorescence imaging and molecular assays. The incorporation of patient-derived dECM enhances translational relevance by better reflecting human bone remodeling and disease progression. This approach provides an accessible, scalable platform for drug testing, regenerative medicine, and mechanistic studies of bone biology. Overall, the protocol offers a practical alternative to high-cost bioprinting techniques and can be readily adapted to other tissue-specific dECM sources.

Introduction

This method aims to establish a cost-effective and reproducible 3D cell culture platform for modelling musculoskeletal diseases, including osteoporosis, by combining GelMA with dECM derived from human bone tissue. This strategy enables the generation of bone-like constructs that more closely replicate the physiological microenvironment than conventional two-dimensional cultures, supporting the study of cell-cell and cell-matrix interactions under under biologically relevant conditions¹. The use of patient-derived osteocytes and clinically sourced dECM from femoral heads enhances translational value and provides a practical link between in vitro experimentation and clinical research needs².

This method was developed to address key limitations of existing 3D culture and biofabrication systems, which often require bioprinters, proprietary biomaterials, or costly consumables (e.g., Matrigel), limiting accessibility accessibility in many^3,4. Moreover, synthetic hydrogels frequently fail to recapitulate the biochemical complexity of mineralized bone extracellular matrix or the behavior of resident osteocytes⁵. In contrast, this protocol incorporates human bone-derived dECM, preserving mineralized content, native proteins, and signaling molecules while avoiding the infrastructure demands associated with additive manufacturing technologies⁶.

The combination of GelMA and dECM provides several advantages. GelMA supports cellular adhesion, viability, and differentiation, and can be processed at 37 °C, enabling cell encapsulation without specialized equipment. Crosslinking using low-cost ultraviolet (UV) sources (365 nm) further increases accessibility⁷. dECM supplementation enriches the construct with tissue-specific cues that are absent in synthetic substitutes, improving osteogenic compatibility and facilitating downstream assays such as immunofluorescence and molecular profiling⁸.

This method aligns with current efforts to develop human-relevant bone models that reduce dependence on animal studies, whose translational accuracy is often limited by interspecies differences^1,9,10. The protocol is consistent with the principles of the 3Rs¹¹, and can be adapted to generate constructs using other tissue-specific dECMs, expanding potential applications to cartilage, skin, and cardiovascular tissue engineering¹².

Overall, this technique provides an accessible and scalable platform for studying bone biology in physiologically relevant conditions. Its compatibility with routine laboratory infrastructure and its capacity to incorporate human clinical samples make it particularly suitable for academic and hospital-based research groups. The method supports not only a platform for further bone studies of bone processes but also could be a starting point for more translational applications in regenerative medicine and personalized therapeutic development.

Protocol

This study was approved by the Scientific Ethics Committee of the Concepción Health Service. Human bone tissue was obtained from femoral heads donated by patients undergoing hip replacement surgery at the Traumatology Hospital of Concepción, Chile, following written informed consent. Samples were stored at −80 °C until use. All procedures complied with institutional guidelines for the ethical use of human tissue in research. All human bone tissue was obtained from femoral heads donated by patients at the Traumatology Hospital of Concepción, Chile, following informed consent and with approval from the Scientific Ethics Committee of the Concepción Health Service. All procedures comply with institutional guidelines for human tissue use.

1. GelMA synthesis

1. Dissolve 10 g of type A gelatin A in 100 mL of phosphate-buffered saline (PBS, pH 7.5) at 60 °C under constant stirring.
2. Reduce the temperature to 50 °C and add 5 mL of methacrylic anhydride at a rate of 0.5 mL/min with stirring gently for 1–3 h.
3. Dilute the reaction mixture 5× with PBS prewarmed at 40 °C to stop the reaction.
4. Dialyze the solution using a 12–14 kDa membrane against 3–5 L of distilled water at 40 °C for 5–7 days, replacing the water daily.
   NOTE: (Critical step) Ensure complete daily water replacement to remove unreacted methacrylic anhydride.
5. Lyophilize the dialyzed solution for 24 h and store the lyophilized GelMA at 4 °C until use.

2. GelMA hydrogel preparation

1. Prepare 10 mL batches of GelMA hydrogels at 5, 10, or 15 w/v%.
2. Dissolve 0.5, 1.0, or 1.5 g of lyophilized GelMA in 9.5 mL of deionized water at 50 °C.
3. Cool the solution to 37 °C.
4. Add 0.5 mL of Irgacure 2959 photoinitiator solution (0.5 w/v% in absolute ethanol).
5. Mix gently and maintain the solution at 37 °C until sterile filtration through a 0.22 µm cellulose acetate membrane filter.

3. Bone decellularization

1. Thaw femoral heads gradually to 25 °C. Cut each femoral head in half and then into smaller pieces.
2. Immerse pieces in sterile distilled water at 60 °C and sonicate for 15 min (40 kHz, continuous).
3. Wash samples in 300 mL of distilled water preheated at 60 °C while shaking at 200 rpm for 5 min.
4. Centrifuge at 1850 × g for 15 min at 25 °C. Repeat the wash and centrifuge steps three times.
5. Sterilize the tissue by sonication (40 kHz, continuous) for 10 min at 60 °C in a solution containing 3% hydrogen peroxide and 0.02% peroxyacetic acid.
6. Transfer the tissue to 300 mL of 70% ethanol and sonicate (40 kHz, continuous) for 10 min at 21 °C.
7. Wash the samples twice with 300 mL distilled water, shake at 200 rpm for 10 min at 60 °C, and centrifuge at 1850 × g for 15 min at 25 °C.
8. First, pre-freeze the bone at -80 °C for 4 h, then lyophilize the decellularized bone for 24 h in a wide-base to ensure even drying and to avoid excessive sample height at 0.05 mbar at -40 to -60 °C.
   NOTE: (Critical step) Ensure the sample does not exceed 2–3 cm in height inside the container to prevent incomplete lyophilization.

4. Bone microparticle preparation

1. Place the decellularized bone fragments in 250 mL Erlenmeyer flasks in the sterilizing solution and homogenize using a homogenizer until a uniform white solution is obtained (Supplementary Video 1).
2. Centrifuge at 1850 × g for 15 min at 25 °C, discard the supernatant, and wash the pellet twice with 300 mL distilled water at 25°C under agitation at 200 rpm for 10 min. Perform a final centrifugation at 1850 × g for 15 min at 25 °C.
3. Lyophilize (0.05 mbar at -40 to -60 °C) the resulting material for 48 h and store at -80 °C until use.
4. Mechanically mill the lyophilized cortical or medullary bone using a sterilized ceramic mortar until fine powder is obtained (Figure 1).
5. Pass the powder through a 200 µm mesh or sieve, weigh the final material, and store at -80 °C.
   NOTE: (Critical step) Ensure the powder is completely dry prior to sieving to prevent aggregation and inaccurate particle size selection.

Supplementary Video 1: Bone homogenization procedure. Homogenization of decellularized bone fragments using a homogenizer until a uniform white suspension is produced.Please click here to download this file.

Figure 1: Bone microparticle preparation. (A) Decellularized bone fragments after homogenization with OV5 homogenizer and storage at -80 °C. (B,C) Sterilized ceramic mortar used to mill decellularized bone into a fine powder. (D) Final decellularized bone microparticles after sieving. Please click here to view a larger version of this figure.

5. Bone gel fabrication

1. Prepare 10 w/v% GelMA by dissolving 1 g of GelMA in 9.5 mL deionized water at 50 °C.
2. Cool the solution to 37 °C and add Irgacure 2959 photoinitiator solution.
3. Maintain the solution at 37 °C and perform sterile filtration using a 0.22 µm filter.
4. Add decellularized bone microparticles (DBMP) at 1% w/v.
5. Mix and pipette up and down carefully, avoiding creating bubbles, at 40 °C until a homogeneous suspension is obtained to form bone gel (BG).
6. Store BG at 4 °C for up to 2 weeks.
   NOTE: (Critical step) Maintain the hydrogel above its gelling temperature (≥37 °C) before cell incorporation to avoid premature solidification.

6. Osteocyte primary culture

1. Bone marrow extraction and initial washing
   1. Extract and mince bone marrow tissue from the interior of the femoral head using sterile surgical instruments (Figure 2).
   2. Transfer the minced tissue to a sterile Petri dish.
   3. Wash the tissue three times with α-MEM culture medium supplemented with antibiotic-antimycotic solution (100 µg/mL penicillin, 100 µg/mL streptomycin, 0.25 µg/mL amphotericin).
   4. Rinse once with Hank’s Balanced Salt Solution (HBSS).
      NOTE: (Critical step) Minimize time between tissue retrieval and processing to preserve cell viability.
2. Enzymatic digestion
   1. Incubate the tissue in collagenase Type I solution for 25 min at 37 °C with gentle agitation.
   2. Wash with HBSS.
   3. Repeat steps 6.2.1–6.2.2 two additional times (three total collagenase digestions).
   4. Incubate the tissue in 0.5 mM ethylenediaminetetraacetic acid (EDTA) for 25 min at 37 °C.
   5. Wash with HBSS and repeat the EDTA step two additional times.
   6. Perform one final collagenase Type I digestion (25 min at 37 °C), followed by an HBSS wash.
      NOTE: (Troubleshooting) Incomplete digestion may result in low cell yield; extend enzymatic incubation by 5–10 min if needed.
3. Cell recovery and culture
   1. Collect the collagenase solution and transfer the final HBSS wash into a 50 mL centrifuge tube.
   2. Centrifuge at 1,000 × g for 5 min at room temperature.
   3. Discard the supernatant and resuspend the pellet in 2 mL of complete α-MEM.
   4. Seed the cell suspension into two T75 culture flasks and incubate at 37 °C with 5% CO₂.
      ​NOTE: Replace culture medium every 2–3 days until adequate confluency (Figure 3) is achieved.

Figure 2: Bone marrow tissue retrieval. (A) Femoral head obtained from hip replacement surgery, showing the medullary region used for cell isolation. (B) Minced medullary bone pieces placed in a sterile Petri dish prior to enzymatic digestion. Please click here to view a larger version of this figure.

Figure 3: Primary osteocyte culture. Osteocytes attached and proliferating in a monolayer within a T75 following isolation from human medullary bone. Please click here to view a larger version of this figure.

7. 3D cell culture

1. Prewarm the amount of BG hydrogel at 37 °C.
   NOTE: (Critical step) Ensure the working area is free of drafts; cooling accelerates premature gelling.
2. Detach cultured osteocytes and centrifuge at 500 × g for 5 min in a 15 mL tube.
3. Resuspend the pellet in 200 µL of culture medium and mix with the BG hydrogel to obtain a final cell density of at least 1000 cells/µL (from now on referred to as bioink).
4. Aspirate 50 µL of the bioink using a 1000 µL pipette and maintain at 37 °C until deposition.
5. Pipette 50 µL of the bioink per construct into each well of the culture plate (Supplementary Video 2); allow 5 min approximately for physical gelation at room temperature (Figure 4).
6. Expose each well to 365 nm UV light for 1 min to achieve irreversible crosslinking (0.30 J/cm²).
7. Add culture medium and incubate the constructs at 37 °C, 5% CO₂ for the required time points. Use a separate plate for each time point.

Figure 4: 3D cell culture construct preparation. Well plate containing cell-laden GelMA constructs immediately after pipetting and following UV (365 nm) crosslinking. Please click here to view a larger version of this figure.

Supplementary Video 2: Preparation of 3D cell-laden constructs. Deposition of hydrogel-cell mixtures into a 12-well plate using a simple pipetting technique prior to UV crosslinking.Please click here to download this file.

8. Biocompatibility testing

1. Live/dead staining
   1. Culture the cell-laden scaffolds for the desired time intervals.
   2. Wash gently with sterile PBS.
   3. Incubate the constructs for 30 min in the dark with live/dead solution containing 0.2% ethidium homodimer-1 (EthD-1) and 0.05% calcein AM in PBS.
   4. Image immediately using a fluorescence microscope with filters for calcein AM (~495/515 nm) and EthD-1 ( ~495/635 nm). Standard fluorescein isothiocyanate (FITC) and Texas Red filter sets are compatible with calcein AM and EthD-1 detection, respectively.
   5. Identify live cells by green cytoplasmic fluorescence and dead cells by red nuclear fluorescence.
      ​NOTE: Planar images can be processed using ImageJ (Fiji). The objective lens, exposure time, gain, and illumination intensity were kept constant for all compared samples and adjusted only to avoid signal saturation. To minimize misinterpretation caused by autofluorescence, the same acquisition settings were maintained for all samples within each experiment, and control observations of scaffold/bone microparticle constructs without cells were used to recognize the background fluorescence pattern associated with the biomaterial. These controls allowed us to distinguish particle-derived signal from true rounded cell staining during image interpretation.
2. Alamarblue quantification
   1. Prepare a 5% alamarblue working solution in complete medium.
   2. Incubate scaffolds in alamarBlue solution for 2.5 h at 37 °C and 5% CO₂, protected from light.
   3. Transfer 100 µL of supernatant to a 96-well plate in triplicate, including controls.
   4. Measure fluorescence at 360/40 nm excitation and 460/40 nm emission.
   5. Normalize fluorescence values to controls to determine relative metabolic activity.

Results

Primary osteocytes were successfully isolated from human femoral heads and were detectable in culture approximately 5 days after extraction. The cells displayed a stellate morphology and formed interconnected networks characteristic of osteocyte morphology, confirming successful isolation, attachment, and viability (Figure 5).

Figure 5: Metabolic activity of 3D osteocyte cultures. Alamarblue metabolic assay for osteocyte-laden constructs at each time interval. Significant differences relative to control are observed across all replicates (ANOVA, P = 0.05). Please click here to view a larger version of this figure.

Overall, the particle size distribution of milled bone was a median of 137.8 (SD ±±38.58). More essential physicochemical characterizations of the prepared GelMA (DoM) and dECM are provided in the Supplemental files (Supplementary Figure 1 and Supplementary Figure 2). Following biofabrication, the 3D bone-like hydrogels maintained high cell viability over time, as demonstrated by qualitative live/dead fluorescence imaging and quantitative metabolic activity measurements using the alamarBlue assay. After seven days of culture, strong calcein-AM fluorescence indicated widespread viable cells throughout the hydrogel matrix, whereas minimal EthD-1 signal confirmed low cytotoxicity (Figure 6).

Figure 6: Live/dead viability assay in 3D constructs. (A1,A2,A3) Viable cells stained with calcein AM (green). (B1,B2,B3) Dead cells stained with ethidium homodimer-1 (red). (C1,C2,C3) Merged images. Each group represents results from three independent experiments (n = 9). Scale bar: 100 µm. Please click here to view a larger version of this figure.

In some preparations, uneven cell distribution or localized regions of reduced viability were observed, particularly in thicker constructs in which nutrient diffusion may be limited. Despite these factors, the method consistently produced constructs with high viability and reproducible cellular network formation across multiple replicates.

Supplementary Figure 1: Protein content and RNA. (A) Donated femoral head with a transversal cut in half. (B) Protein content (absorbance). (C) RNA measurements as a function of subsequent washing steps during the decellularization process. (D) RNA ran on an agarose gel.Please click here to download this file.

Supplementary Figure 2: ¹H-Nuclear magnetic resonance. (A) Gelatin precusor and (B) GelMA for quantification of the degree of methacrylation. 1H-Nuclear magnetic resonance of Gelatin precursor and GelMA for quantification of the degree of methacrylation. Degree of methacrylation (DoM) of ~66 % was calculated, depicting aromatic amino acids resonances (7.24-7.51 ppm) that were not chemically modified and green lysine amino acids (3.0–3.1 ppm). Methacrylate vinyl groups that appear after reaction (5.3–5.6 ppm).Please click here to download this file.

Discussion

The presented protocol offers an accessible and translational method for generating 3D bone-like constructs, addressing key limitations of current approaches such as reliance on expensive bioprinters or proprietary matrices¹³. By combining GelMA with human decellularized bone extracellular matrix⁸, physiologically relevant constructs can be fabricated using only standard laboratory equipment. This simplicity enables adoption in resource-limited environments while preserving essential cues for studying musculoskeletal disease mechanism. The use of primary human osteocytes¹⁴ and clinical bone tissue further strengthens translational applicability by supporting patient-relevant modeling of bone disorders and evaluation of biomaterial responses under near-physiological conditions^6,15. Using this platform as a foundation, more targeted studies can be developed to investigate defined biological readouts, such as drug response in osteoporotic-derived cells, changes in osteocyte-specific gene expression profiles, and modulation of key markers associated with bone remodeling (e.g., sclerostin or RANKL).

Several critical steps are essential to ensure reproducibility and material performance. During GelMA synthesis, complete daily water replacement is required to remove residual methacrylic anhydride, as incomplete purification may affect crosslinking efficiency and cytocompatibility. For lyophilization, sample height should not exceed 2–3 cm to ensure uniform drying and prevent residual moisture, which can compromise subsequent processing. Prior to sieving, the decellularized bone powder must be completely dry, as residual moisture promotes aggregation and leads to inaccurate particle size distribution. During bioink preparation, the hydrogel must be maintained above its gelling temperature and the plastics associated with their handling (≥37 °C) to avoid premature solidification and ensure homogeneous cell distribution. Additionally, minimizing the time between tissue retrieval and processing is critical to preserve cell viability, particularly when working with primary human osteocytes.

From a technical standpoint, modifications and troubleshooting strategies can improve consistency and data interpretation. Imaging of dense and mineralized 3D constructs remains challenging due to optical scattering and depth limitations inherent to hydrogel-based systems^8,16. To address this, standardized acquisition parameters (objective, exposure time, gain, and illumination intensity) were maintained across samples, with adjustments only to avoid signal saturation. Planar image analysis can be performed using ImageJ (Fiji) to enable consistent processing. Given the presence of bone-derived particles, autofluorescence may occur in both live/dead channels; thus, scaffold-only controls were systematically used to identify background signal patterns and to differentiate between rounded cells immersed in the construct. This approach allowed reliable distinction between particle-associated fluorescence and true cell-associated staining, based on morphology and spatial distribution, ensuring accurate qualitative assessment of cell viability.

Despite its advantages, several limitations should be considered. Donor variability in bone matrix composition may influence scaffold performance and cell behavior. As terminally differentiated cells, osteocytes exhibit limited proliferative capacity, which may restrict long-term expansion, particularly when derived from osteoporotic tissue. In addition, the present study does not include molecular validation of osteocyte identity through specific markers or functional readouts, which should be considered in future work to further confirm cell phenotype and remodeling activity. Imaging and quantification within dense 3D hydrogels remain challenging due to optical scattering and depth-related limitations. These issues underscore the need for standardized handling, thin constructs when possible, and improved imaging strategies^8,16, such as several planes registered by hyperstack confocal acquisition.

Compared to traditional 3D culture and bioprinting systems, this method offers a low-cost alternative that retains high biological fidelity while avoiding specialized instrumentation. Simple pipetting yields constructs with consistent viability and cellular distribution to obtain osteocyte viability, network formation, and compatibility with downstream assays. The biomimetic environment created by combining GelMA with bone-derived dECM enables further studies of bone remodeling, osteogenesis, mechanotransduction, and biomaterial interactions. Additionally, the workflow can be adapted to other tissue-specific dECMs, broadening potential applications in regenerative medicine and disease modeling, particularly in hospital-based research settings.

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
2-Hydroxy-4′-(2-hydroxyethoxy)-2-methylpropiophenone (Irgacure 2959)	Sigma-Aldrich	410896-10G	Hydrogel polymerization photoinitiator
AlamarBlue cell viability reagent	Invitrogen	DAL1100	For quantitative viability assay
Antibiotic-Antymycotic (100´)	Gibco	15240-062	Culture medium component
Cell culture flask (T75)	SPL Life Sciences	70075	For osteocyte culture
Collagenase Type 1	Gibco	17100-017	For enzymatic digestion of bone samples
Conical tube, 15 mL	SPL Life Sciences	50015	For Hydrogel preparation
DarkDawn D15-365 nm	DarkDawn	-	For UV Crosslinking of hydrogel
DPBS, powder, no calcium, no magnesium	gibco	21600069	Hydrogel solvent, LIVE/DEAD kit solvent
EDTA Disodium Salt 2-HID	Winkler	ED-0760	For enzymatic digestion of bone samples
Eppendorf centrifuge 5702	Eppendorf	5702000010	For bone microparticle preparation
Ethanol absolute for analysis	Supelco	1,00,98,32,511	For bone descellularization (diluted to 70%), photoinitiator solvent
Fetal bovine serum	Gibco	A52567-01	Culture medium component
Gelatin from porcine skin	Sigma-Aldrich	G1890-500G	Hydrogel component
HBSS (10´)	Gibco	14185-052	For washing of bone marrow samples
L-Ascorbic acid 2-phospate	Sigma-Aldrich	A8960-5G	Culture medium component
LIVE/DEAD viability/cytotoxicity Kit	Invitrogen	L3224	For qualitative viability/cytotoxicity assay
MEM Alpha (1´)	Gibco	12571-063	Culture medium component
Methacrylic anhydride	Sigma-Aldrich	276685-500G	Hydrogel component
Micro centrifuge tube	Citotest	4610-1844	For preparation of alamarBlue and LIVE/DEAD dyes
NuncIon delta surface	Thermoscientific	150628	12 well plate for seeding of samples
OLYMPUS BX43 fluorescence light source Olympus U- LH100HG	Evident	U- LH100HG	Fluorescence analysis light source
RC-6 Plus centrifuge	Thermo Scientific Sorvall	46910	For bone descellularization and bone microparticle preparation
Synergy HTX multi mode reader	BioTek Instruments	-	Spectrophotometer
Syringe with needle	NIPRO	SY3-WN-R-03	For Hydrogel filtering
Tissue culture plate 96 well	Jet Biofil	TCP001096	96 well for alamarBlue viability assay.
Ultrasonic cleaner	VWE	97043-958	For sonication of bone samples