Engineering a cell-free bone regeneration platform using osteogenically primed MSC-EVs and nHAp-enriched IPN hydrogels

ABSTRACT

Aims

This study aimed to enhance the osteoinductive potential of mesenchymal stem cell-derived extracellular vesicles (MSC-EVs) by integrating them into a nano-hydroxyapatite (nHAp)-enriched hydrogel scaffold for bone regeneration applications.

Materials & methods

EVs were isolated from naïve and osteogenically primed MSCs and characterized for morphology, cargo content, and cytocompatibility. Their uptake and osteoinductive activity were assessed in vitro using MC3T3 cells within a 3D interpenetrating network (IPN) hydrogel. The most effective EV formulation was incorporated into an nHAp – IPN hydrogel scaffold and evaluated both in vitro and in a murine subcutaneous implantation model.

Results

Primed MSC-EVs showed elevated calcium, ALP activity, and osteogenic/angiogenic mRNAs (Runx2, Vegf-a) compared to naïve EVs, with comparable size and morphology. Both EV types were internalized efficiently without cytotoxicity. In combination with nHAp, primed EVs enhanced ALP activity, calcium deposition, and in vivo mineralization. Histological analysis confirmed scaffold biocompatibility and mineralized tissue formation.

Conclusions

Osteogenically primed MSC-EVs significantly improved the osteoinductive performance of nHAp-based hydrogels, supporting their potential as a cell-free therapeutic strategy for bone tissue engineering.

Plain Language Summary

Medical procedures like bone transplants or stem cell therapy are frequently necessary for bone injuries or deformities that don’t repair properly. These approaches, however, may have disadvantages, such as the possibility of rejection or complicated processes. Researchers are looking at safer, “cell-free” alternatives that employ natural chemicals rather than biological cells. Extracellular vesicles (EVs), which are microscopic particles that are spontaneously produced by stem cells and contain crucial signals to aid in tissue regeneration, are one such alternative. In this work, EVs were extracted from stem cells that had undergone specific treatment to promote bone formation. The ability of these EVs to contribute in the formation of new bone was then examined in laboratory tests and in a mouse model. This work used a 3D gel-like material in conjunction with nano-hydroxyapatite, a mineral that resembles bone, to deliver the EVs. The findings demonstrated that EVs derived from treated stem cells were more effective than untreated EVs at promoting cell growth and the formation of bone-like tissue. Crucially, the EVs did not damage the nearby cells, and the material was harmless. According to this study, integrating engineered EVs with bone-like materials may offer a viable and secure method of bone repair that eliminates the need for conventional transplants or living cells.

GRAPHICAL ABSTRACT

1. Introduction

The increasing incidence of traumatic bone injuries and age-related degenerative conditions, such as osteoporosis has intensified the search for innovative and effective strategies to repair bone defects. Traditional approaches, such as autografts and allografts, are widely used but are often hindered by limitations including donor site morbidity, risk of immune rejection, and high procedural costs [1,2]. While cell-based therapies involving mesenchymal stem cells (MSCs) have demonstrated regenerative promise, their clinical application remains constrained by issues such as unpredictable differentiation, ethical concerns, and complex handling and storage requirements. This has led to growing interest in cell-free regenerative strategies, particularly those harnessing the paracrine potential of MSCs through their secreted extracellular vesicles (EVs) [3,4].

EVs, including microvesicles and exosomes, are nanoscale vesicles released by cells that play a crucial role in intercellular communication [5]. In the context of bone regeneration, EVs derived from osteogenically committed or primed cells carry a range of osteoinductive factors capable of stimulating lineage-specific differentiation in target cells [5]. Among the various model systems used to study osteogenesis, the MC3T3-E1 preosteoblast cell line is widely accepted for its osteoblast-like behavior, including proliferation, matrix production, and mineralization. One of the major challenges in EV-based therapies is the delivery method [6,7]. Direct injection of EVs to injury sites typically requires repeated dosing due to their short retention time. To address this, hydrogels, particularly those based on natural polymers like alginate, have been explored as delivery matrices for sustained EV release [7]. Alginate hydrogels offer biocompatibility, ease of gelation, and cargo-protective properties, enabling localized and prolonged exposure of therapeutic vesicles to surrounding tissue. Moreover, the transition from two-dimensional (2D) systems to three-dimensional (3D) hydrogel-based microenvironments more accurately replicates the architecture of native tissue [8]. This transition enhances cell – cell and cell – matrix interactions and facilitates the functional integration of EV-mediated signals [9]. However, most polymeric hydrogels alone lack the mechanical strength and biological complexity required for optimal bone regeneration. To enhance structural integrity and simulate the mineralized environment of native bone, researchers have incorporated bioactive ceramics such as nano-hydroxyapatite (nHAp) into hydrogel systems [10,11]. nHAp, which makes up the majority of bone’s mineral content, provides osteoconductive surfaces, supports cell adhesion, and promotes differentiation. However, due to its brittle nature, nHAp must be combined with other polymers for practical scaffold construction [12].

As an innovative approach to overcome this limitation, our previous work involved the use of interpenetrating polymer networks (IPNs), in which a synthetic polymer such as polyethylene glycol diacrylate (PEGDA) is crosslinked within a natural alginate matrix, substantially improving mechanical resilience while preserving cellular compatibility [12]. Building on this strategy, the present study developed a biofunctional scaffold composed of an alginate – PEGDA IPN, enhanced with nano-hydroxyapatite to recreate a mineralized microenvironment, and primed MSC-derived EVs to supply osteoinductive signaling [12]. Previously standardized protocols were used to optimize EV concentration and hydrogel formulation [13]. The study first evaluated the bioactivity of EVs isolated from both naïve and osteogenically primed MSCs, focusing on their influence on preosteoblast differentiation and bone-associated molecular markers. The osteoinductive potential of these vesicles was then tested in combination with nHAp within the hydrogel system to explore synergistic effects.

In recent years, MSC-EVs have gained attention as a promising tool for bone regeneration due to their role in mediating cell-cell communication and delivering bioactive molecules such as miRNAs, proteins, and lipids [9,14]. Several studies have demonstrated the ability of naïve MSC-EVs to support osteogenic differentiation and matrix mineralization. For example, Wang et al. highlighted the role of MSC-EVs in promoting bone repair by modulating the local microenvironment and enhancing osteoblast function [15]. Similarly, Kuang et al. reported that MSC-EVs carrying specific miRNAs can influence signaling pathways associated with osteogenesis [16]. While these findings are encouraging, most current research utilizes EVs from unstimulated MSCs, which may limit their regenerative potential. Moreover, many studies do not thoroughly investigate the molecular cargo of EVs or their functional interactions with osteoconductive scaffolds such as nHAp. The long-term in vivo performance and integration of EV-loaded scaffolds also remain underexplored. Pishavar et al. noted a lack of standardization in EV isolation and scaffold integration strategies, which hinders reproducibility and translational progress [17]. These limitations underscore the importance of developing optimized EV preparations, such as those derived from osteogenically primed MSCs, and exploring their synergistic effects with bone-mimicking biomaterials to enhance regenerative outcomes.

To validate the scaffold’s performance in vivo, a subcutaneous pouch assay in C57BL/6 mice was conducted. Hydrogels with different combinations of EVs and nHAp were implanted, and outcomes were assessed via histology and Alizarin Red staining. Notably, the absence of an inflammatory response across all groups, along with increased mineral deposition in constructs that combined EVs with mineralized components, confirmed the biocompatibility and osteogenic potential of the scaffold. These findings suggest that this cell-free, dual-functionalized hydrogel system could serve as a promising platform for bone tissue regeneration, offering a viable alternative to existing cell- and graft-based therapies.

This study explored a novel, cell-free bone regeneration strategy by integrating osteogenically primed MSC-derived extracellular vesicles (EVs) into a nano-hydroxyapatite – reinforced IPN hydrogel scaffold. Unlike previous studies using naïve EVs or single network hydrogels, this platform enhances osteogenesis through synergistic effects, demonstrated in vitro and in vivo. It addresses key gaps in EV delivery and scaffold performance, offering a safer, scalable alternative to cell-based therapies with strong potential for clinical translation.

2. Materials and methods

2.1. Materials

Materials were procured from a range of suppliers. Fetal bovine serum (FBS) was sourced from Gibco (USA), while glutamine, penicillin-streptomycin, and α-minimal essential medium (α-MEM) were obtained from HiMedia (India). Sigma-Aldrich (Germany) supplied several chemicals, including sodium alginate (derived from brown algae, with G-block: 14.0 − 31.0% and molecular weight: 32,000 − 400,000 g/mol), calcium sulfate, glutaraldehyde, β-glycerophosphate, ascorbate-2-phosphate, dexamethasone, insulin transferrin-selenium (ITS) Premix, ascorbic acid, sodium pyruvate, phosphatase substrate, nHAp ( < 200 nm particle size), and sulfuric acid. Polyethylene glycol diacrylate (PEGDA) was acquired from SUNBIO (Korea). Invitrogen (USA) provided the PicoGreen double-stranded DNA quantification and live/dead assay kits. The QuantiChrom calcium assay kit was procured from Bioassay Systems (USA). For molecular biology applications, the PrimeScript™ 1st strand cDNA synthesis kit was obtained from Takara (Japan), and the PowerUp™ SYBR™ Green Master Mix was obtained from Applied Biosystems (USA).

2.2. Cell culture and extracellular vesicle isolation

The pre-osteoblast cell line MC3T3-E1 (hereafter referred to as MC3T3) was purchased from ATCC (ATCC CRL-2593). Five million cells were used in each experiment as target cells and were maintained in α-MEM with 10% FBS throughout the investigation. Mouse MSCs were used as an EV source under primed and naïve conditions. MSCs were harvested from the bone marrow of 5–6-week-old mice (C57BL/6, procured from ACTREC, Mumbai) and maintained on α-MEM supplemented with 1% penicillin/streptomycin antibiotic and 20% FBS. All animal procedures were approved by the Symbiosis International University’s Institutional Animal Ethics Committee (IAEC) [SSBS/IAEC/02–2019]. MSCs at passage 3 were used throughout the experiments. EVs were isolated at two stages from naïve MSCs and MSCs primed with osteogenic differentiation medium. For EV isolation, MSCs at 80–90% confluency were serum starved for 48 hours, and then EV extraction was done from the conditioned medium (CM). CM was collected and centrifuged at 2600 g for 10 minutes at 4°C to remove cell debris and apoptotic bodies. To isolate total EVs, the apoptotic body-depleted supernatant was centrifuged at 100,000 ×g for 70 minutes. The protein content of total EVs was determined using the bicinchoninic acid (BCA) assay. EVs were isolated using a standardized protocol previously established Kulkarni et al. [18], employing the same mouse MSC source and conditions. Comprehensive EV characterization, including dynamic light scattering (DLS) and Western blotting for positive and negative markers, was performed in that study and thus not repeated here.

2.3. Osteogenic and chondrogenic differentiation of MSCs

The potential for tri-lineage differentiation of isolated MSCs was investigated by culturing the MSCs in two specific differentiation media. For osteogenic differentiation, the medium included 10 nM dexamethasone, 10 mM β-glycerophosphate, and 50 μg/mL ascorbate-2-phosphate, all within high-glucose DMEM supplemented with 10% FBS [19]. For chondrogenic differentiation, the medium comprised 100 nM dexamethasone, 1% insulin transferrin-selenium (ITS)-Premix, 1 μg/mL ascorbic acid, 1% sodium pyruvate, and 10 ng/mL Human TGF-β1, also in high-glucose DMEM with 10% FBS [20]. These conditions were maintained for 14 days to evaluate the differentiation potential. Osteogenic differentiation was confirmed by the presence of calcium deposits, as indicated by Alizarin Red staining. In contrast, chondrogenic differentiation was assessed through Safranin O staining, which detects acidic proteoglycans in chondrocytes after the 14-day culture period.

2.4. Osteogenic priming of MSCs and preparation of conditioned medium for EV isolation

Mouse bone marrow-derived MSCs, at passage 3, were cultured in an osteogenic induction medium to generate primed EVs. The induction medium was composed of high-glucose DMEM supplemented with 10% FBS, 10 nM dexamethasone, 10 mM β-glycerophosphate, and 50 μg/mL ascorbate-2-phosphate. MSCs were maintained under these conditions for a total of 14 days to induce osteogenic differentiation. Once cells reached 80–90% confluence, they were washed and maintained in serum-free medium for 48 hours. The resulting CM was collected and processed for EV isolation. This priming strategy was used to enhance the osteoinductive potential of EV cargo. The presence of calcium-rich deposits following the priming period was confirmed by Alizarin Red staining. For comparison, naïve EVs were obtained from MSCs cultured in standard growth medium (high-glucose DMEM supplemented with 10% FBS), followed by the same serum starvation and CM collection protocol. Further details of EV isolation are provided in Section 2.2, and multipotency validation, including chondrogenic differentiation, is described in Section 2.3.

2.5. Live-dead assay

The cytotoxicity of the MSC-EVs and mineralized microenvironment was evaluated using a cell viability assay on days 1, 7, and 14. Cellular hydrogel discs (n = 3) were subjected to a dual-fluorescence staining protocol. Horizontal sections (1 mm thick) were incubated with ethidium homodimer-1 (4 mM) and calcein-AM (2 mM) following the manufacturer’s instructions. After washing with PBS to remove excess dye, the samples were analyzed using fluorescence microscopy. Image acquisition was performed using excitation and emission wavelengths of 528 nm and 617 nm, respectively. Z-stack images were obtained using 1 × 1 binning mode. Quantification of cell viability was performed through manual cell counting, distinguishing between viable (green fluorescence) and non-viable (red fluorescence) cells using image analysis software. The percentage of viable cells was then calculated based on these counts.

2.6. Biochemical assays

Biochemical analyses were conducted on all experimental groups (n = 3) at days 1, 7, and 14. Samples were extracted, washed with phosphate-buffered saline (PBS), and mechanically disrupted in microcentrifuge tubes using a pestle. The resulting fragments were suspended in 300 μL of passive lysis buffer (Promega, Madison, WI, USA). Following a freeze-thaw cycle, the lysates were briefly sonicated (Sonics, Vibra Cell, USA). Centrifugation at 10,000 rpm for 5 minutes separated the supernatant from the pellet, with the supernatant stored at −20°C for subsequent analysis. Deoxyribonucleic acid (DNA) content and intracellular alkaline phosphatase (ALP) levels were quantified from the supernatant. DNA quantification was performed using a PicoGreen double-stranded DNA assay kit (Invitrogen, USA), and concentrations were determined using a standard curve. Intracellular ALP content was determined via a p-nitrophenyl phosphate colorimetric assay, measured at 405 nm using a microplate reader (BioTek Microplate Reader H1M, GEN5 V3.03, USA) [21]. ALP activity was normalized by dividing ALP units by DNA content (in mg). To assess mineralization, sample pellets were incubated overnight at 37°C in 0.9N sulfuric acid to dissolve surface calcium deposits. Calcium deposition in the seeded hydrogel scaffolds was quantified using a calcium assay kit (QuantiChrom calcium assay kit, Bioassay Systems, USA). Acellular discs with equivalent hydroxyapatite concentrations or residual calcium from alginate IPN discs served as controls. These controls were used to account for calcium content in the mineralized microenvironment and alginate hydrogel [21].

2.7. Characterization of EVs

2.7.1. Morphology

Field emission scanning electron microscopy (FESEM) (Photometrics Inc., USA) was employed to assess the morphological characteristics and size of EVs. The EVs were subjected to a washing procedure using 1× PBS and subsequently dispersed uniformly on Silicon wafer substrates (Global Nanotech). Following air-drying, the samples underwent fixation with 2.5% glutaraldehyde solution for 30 minutes at ambient temperature. The fixed specimens were then rinsed thrice with sodium phosphate buffer. A graded dehydration protocol was implemented, utilizing ethanol solutions of increasing concentrations (10%, 30%, 50%, 70%, 80%, 90%, and 100%) for 10-minute intervals each. The dehydrated samples were allowed to air-dry at room temperature. The prepared specimens were then mounted onto aluminum stubs, sputter-coated with gold, and subjected to FESEM analysis for detailed structural examination.

2.7.2. Examination of EVs cargo

The presence of ALP enzyme in EVs was investigated using a p-nitrophenyl phosphate colorimetric assay [13]. A total of 180 μg of EV content underwent lysis through a combination of passive lysis buffer incubation (45 minutes on ice) and physical disruption via syringe. The resulting lysate was subjected to 50 mM p-nitrophenyl phosphate substrate treatment and incubated at 37°C for 40 minutes. Absorbance measurements were recorded at 405 nm, with passive lysis buffer serving as a blank. ALP activity was quantified using the equation: ALP units/mL = A405/(18.5 x incubation time in minutes x sample volume in mL). To assess calcium content, 180 μg of EVs were lysed following the aforementioned procedure. Calcium quantification was performed using a commercial assay kit, with results normalized to EV protein concentration. The presence of osteogenesis and angiogenesis-associated mRNAs in EVs was evaluated using a TRIzol® reagent-based method. RNA purity and quantity were determined using a NanoDrop. Complementary DNA (cDNA) was synthesized using a commercial kit. Real-time PCR analysis was conducted to assess the expression of the osteogenic marker Runx2 and the angiogenic marker Vegfa, with Gapdh serving as a housekeeping gene (Table 1). PCR reactions were performed using a SYBR Green-based master mix, and cycle threshold (Ct) values were recorded for each gene of interest.

Table 1.

Gene-specific primer sequence used for real time-PCR.

Genes	Gene ID	Primers	Sequences
Gapdh	NM_001289726.1	Forward	5’-ACTGCCACCCAGAAGACTGT-3´
		Reverse	5’-CCATGCCAGTGAGCTTCC-3’
Runx2	NM_001271630.1	Forward	5’-GCCGGGAATGATGAGAACT-3’
		Reverse	5’-GGACCGTCCACTGTCACTTT-3’
Vegf-a	NM_001025250.3	Forward	5’-GGAGTACCCCGACGAGATAG-3’
		Reverse	5’-CTATGTGCTGGCTTTGGTGA-3’

2.7.3. Internalization of MSC-EVs by MC3T3 cells

MSC-EVs were fluorescently labeled using Dil dye according to the manufacturer’s specifications. A concentration of 20 μg/mL of labeled EVs was introduced to MC3T3-E1 cells, which were subsequently seeded onto coverslips to facilitate adherence [13]. The temporal dynamics of EV uptake were assessed at 1, 24, 48, and 72 hours post-introduction using an immunofluorescence microscope (Carl Zeiss Microscope Apotome HXP 120, Germany). Cellular nuclei were counterstained with DAPI to provide context for the localization of internalized EVs within the cellular architecture.

2.8. Screening of MSC-derived EVs in naïve and primed conditions for enhanced osteoinductive potential

MSC-derived EVs, obtained under both naïve and osteogenically primed conditions, were encapsulated within alginate hydrogels at a concentration of 30 μg/mL. This concentration was selected based on results from our previous experiments, where it demonstrated optimal bioactivity [13]. MC3T3 pre-osteoblast target cells were co-encapsulated within the hydrogel matrix to evaluate the osteoinductive effects of the embedded EVs under standardized 3D culture conditions. The synthesis of the alginate hydrogel followed our previously established and published protocol [7]. Briefly, a sterile 2.5% (w/v) sodium alginate solution was prepared in α-MEM and filtered using a 0.22 μm membrane filter. A calcium sulfate (CaSO₄) slurry (8.4 g in 40 mL distilled water) was used as the ionic crosslinker. The alginate solution (800 μL) was mixed with 160 μL of α-MEM containing 40 μL of CaSO₄ slurry by passing the mixture between two 1 mL syringes connected via a three-way stopcock, ensuring homogeneity. The final hydrogel concentration was adjusted to 2.0% (w/v) alginate. Following the incorporation of EVs and MC3T3 cells, the solution was cast into silicone molds (5 mm diameter × 2 mm thickness), sandwiched between sterile glass slides, and incubated at 37°C for 40 minutes to facilitate complete crosslinking. These EV-loaded hydrogel constructs were then utilized for subsequent assays to compare the osteogenic potential of EVs collected under naïve and osteogenically primed conditions. The various experimental and control groups utilized for the investigation are described in Table 2.

Table 2.

Description of various experimental and control groups used in evaluating the osteoinductive potential of MSC-EVs collected in naïve and osteogenically primed conditions.

Group	Details
Cells control (control group)	MC3T3 cells encapsulated in alginate
Cells+ MSC Naïve EVs	MC3T3 cells encapsulated along with 30 µg/mL naïve MSC-EVs in alginate
Cells+ MSC Osteo EVs	MC3T3 cells encapsulated along with 30 µg/mL primed MSC-EVs in alginate

2.9. Synergistic effect of primed MSC-derived EVs and mineralized microenvironment in IPN hydrogel

The formulation of the biomineralized IPN hydrogel and the concentration of EVs used in this study were adapted from our previously published work [12,13]. In brief, a mineralized microenvironment was created within the first sodium alginate network by directly incorporating nano-hydroxyapatite (nHAp) at a concentration of 5 mg/mL. EVs derived from osteogenically primed MSCs were co-encapsulated at a concentration of 90 μg/mL along with five million MC3T3 pre-osteoblast cells per construct. Following alginate gelation, the resulting discs were incubated in a 15% (w/v) PEGDA solution for 2.5 hours to allow diffusion of PEGDA into the matrix. The second network was formed the following day by photopolymerizing PEGDA, thereby completing the IPN structure. Photopolymerization of the secondary PEGDA network was performed using UV light at 312 nm for 10 minutes total (5 minutes per side), ensuring consistent crosslinking across all hydrogel batches. Constructs lacking nHAp were prepared in parallel and served as non-mineralized controls to isolate the effect of the biomineralized microenvironment. All IPN hydrogels were maintained in α-MEM supplemented with 10% FBS under standard culture conditions. Samples were collected at days 1, 7, and 14 for downstream biochemical and molecular analyses. Osteogenic differentiation was evaluated by measuring normalized alkaline phosphatase (ALP) activity and calcium deposition across the specified time points. The various experimental and control groups utilized for the investigation are described in Table 3.

Table 3.

Description of various experimental and control groups used in evaluating the osteoinductive potential of primed MSC-EVs and mineralized microenvironment.

Group	Details
Cells control (control group)	MC3T3 cells encapsulated in alginate-PEGDA IPN
Cells + nHAp powder (nHAp control)	MC3T3 cells encapsulated along with 5 mg/mL nHAp powder in alginate-PEGDA IPN
Cells + MSC-EVs (EV control)	MC3T3 cells encapsulated along with 90 µg/mL primed MSC-EVs in alginate-PEGDA IPN
Cells+nHAp powder + MSC-EVs	MC3T3 cells encapsulated along with 5 mg/mL nHAp powder and 90 µg/mL primed MSC-EVs in alginate-PEGDA IPN

2.10. In vivo pouch assay in C57BL/6 mice for evaluating osteoinductive potential of mineralized IPN hydrogels with encapsulated EVs

An in vivo subcutaneous pouch assay was conducted in C57BL/6 mice to evaluate the osteoinductive potential of mineralized IPN hydrogels containing MSC-derived EVs. All animal procedures were approved by the Symbiosis International University Institutional Animal Ethics Committee (IAEC) [SSBS/IAEC/04–2021]. A total of 20 mice (4–8 weeks old, 20–30 g, female) were randomly divided into four experimental groups (n = 5 per group): (I) alginate – PEGDA hydrogel, (II) alginate – PEGDA with a mineralized microenvironment (nano-hydroxyapatite, 5 mg/mL), (III) alginate – PEGDA with EVs (90 μg/mL), and (IV) alginate – PEGDA containing both a mineralized microenvironment and EVs (Table 4). All hydrogel constructs were acellular and were prepared by encapsulating the respective components into the alginate network prior to PEGDA diffusion and photopolymerization, as established in our prior work [12]. Under anesthesia, a 10 mm dorsal incision was made over the upper thoracic spine of each mouse, and a single hydrogel construct was implanted subcutaneously into a surgically created pouch followed by closure with absorbable sutures. Postoperative care was provided in accordance with institutional animal ethics guidelines. Animals were housed at a temperature of 21 ± 1°C and humidity of 45–55% in polycarbonate cages alternating 12 h light/dark cycle. After a 4-week implantation period, mice were euthanized by sedation with isoflurane followed by cervical dislocation, in accordance with institutional animal care guidelines. The implants along with surrounding tissues were harvested. Histological analysis, including hematoxylin and eosin (H & E) staining and Alizarin Red S staining, was performed to assess tissue integration, new bone formation, and osteogenic marker expression. The group size (n = 5) was based on prior pilot data and power analysis, ensuring sufficient statistical power while minimizing animal use. This study was conducted in accordance with the ARRIVE (Animal Research: Reporting of In Vivo Experiments) guidelines.

Table 4.

Animal experimental group details.

Groups	Experimental groups
Group I, n = 5	Alginate-PEGDA IPN
Group II, n = 5	Alginate-PEGDA IPN with mineralized microenvironment (5 mg/mL nHAp)
Group III, n = 5	Alginate-PEGDA IPN with encapsulated Primed MSC-EVs (90 μg/mL)
Group IV, n = 5	Alginate-PEGDA IPN with both mineralized microenvironment (5 mg/mL nHAp) and primed MSC-EVs (90 μg/mL)

2.11. Statistical analysis

The data presented in the graphs are mean ± standard deviation (SD) for three biological replicates (n = 3). GraphPad Prism 8 (GraphPad Software, San Diego, USA) was used for statistical analysis. For DNA, normalized ALP, and calcium measurements, two-way analysis of variance (ANOVA) was performed with correction for multiple comparisons, using a significance threshold of < 0.05.

3. Results

3.1. Tri-lineage differentiation of MSCs

MSCs can potentially differentiate into osteoblasts, chondrocytes, and adipocytes [9]. It is advisable to prove at least two lineages of directed differentiation of MSCs, as it can prove their multipotency. The current study provided the differentiation ability of isolated MSCs into chondrogenic (Figure 1(A)) and osteogenic (Figure 1(B)) lineages.

Figure 1.

Characterization of MSCs and their EVs. (A) safranin O staining of MSC pellets cultured for 14 days in standard growth medium (left) and chondrogenic induction medium (right) demonstrates chondrogenic differentiation. (B) alizarin red S staining of MSCs cultured for 14 days in growth medium (left) and osteogenic induction medium (right) indicates mineralized matrix deposition as evidence of osteogenic differentiation. (C) FESEM images of EVs isolated from MSCs under naïve (left) and osteogenically primed (right) conditions show comparable spherical morphology and size distribution. (D) uptake of Dil-labeled naïve (left) and primed (right) MSC-EVs by MC3T3 cells after 24 hours of incubation, visualized via fluorescence microscopy. Nuclei were counterstained with DAPI (blue), and internalized EVs appear as red puncta, confirming cellular internalization. (E) comparative gene expression profiling of osteogenic (Runx2) and angiogenic markers (Vegfa) in naïve and osteogenically primed MSC-derived EVs. Scale bars; 20 µm (A, B and D). Data are presented as mean ± SD (n = 3, *p < 0.05, **p < 0.01).

Chondrogenic differentiation of MSCs was confirmed by safranin O stain, which stains proteoglycans of chondrocytes. MSCs, when cultured in a growth medium (Figure 1(A), left), serve as a negative control. In contrast, MSCs cultured with a chondrogenic differentiation medium showed positive staining, suggesting the chondrogenic differentiation ability of MSCs (Figure 1(A), right). Mineralization is a marker for osteogenesis. Here, after 14 days of culture in growth media, calcium secretion (mineralization) was evaluated by Alizarin Red S staining (Figure 1(B)), left) and osteogenic differentiation medium (Figure 1(B), right). Cells were stained red as positive staining for osteogenic differentiation, unlike in growth medium, due to the binding of calcium ions in the cell with alizarin dye as a sign of differentiation of MSCs into osteoblasts, confirming the ability of MSCs to differentiate into osteoblasts.

3.2. Characterization of MSC-EVs

3.2.1. Morphology study for MSC-EVs using FESEM

FESEM was used to characterize EVs isolated from MSCs cultured under naïve and osteogenically induced conditions (Figure 1(C)). The EVs displayed typical round to ovoid morphology. No observable differences in size, shape, or surface structure were noted between the two groups. These results indicate that osteogenic priming does not affect the gross morphology of MSC-derived EVs.

3.2.2. Uptake of Dil-labeled MSC-EVs by MC3T3 cells in 2D culture

To assess the uptake of MSC-derived EVs, Dil-labeled EVs were co-cultured with MC3T3 cells on coverslips and monitored over 72 hours (Figure 1(D)). Fluorescent imaging revealed progressive internalization beginning as early as 1 hour post-seeding and continuing up to 72 hours. Both naïve and osteogenically primed EVs were internalized by MC3T3 cells, as evident from the intracellular Dil signal localization. At the 24-hour time point, substantial uptake of EVs from both sources was observed, confirming the ability of MC3T3 cells to internalize MSC-derived EVs regardless of their origin (Figure 1(D)). These findings support the suitability of this model for evaluating EV-mediated cargo delivery [22,23].

3.2.3. Cargo profiling of naïve and osteogenically primed MSC-derived EVs

To investigate the bioactive cargo content of MSC-derived EVs, a comparative analysis was performed between vesicles derived from naïve and osteogenically primed MSCs. Calcium was detected exclusively in EVs from primed MSCs, while it was absent in those from naïve cells. Both EV types carried ALP; however, the levels were significantly elevated in primed EVs, showing a 2.82-fold increase relative to naïve counterparts. Furthermore, the expression of osteogenesis-related mRNA, Runx2, was 3.48-fold higher in primed EVs (Figure 1(E)). Similarly, angiogenesis-associated mRNA, Vegfa, was upregulated by 2.49-fold in primed EVs compared to naïve ones (Figure 1(E)). These results highlight the enhanced osteoinductive and angiogenic cargo profile of primed MSC-EVs, supporting their potential utility as cell-free therapeutic agents in regenerative medicine.

3.2.4. Target cell viability and osteoinductive potential of MSC-EVs

EVs obtained from both naïve and osteogenically primed MSCs were first assessed for cytotoxicity and their ability to promote osteogenic differentiation in target cells. Cytotoxicity analysis revealed no adverse effects from either naïve or primed MSC-EVs over a 14-day incubation period (Figure 2(A)). In fact, a significant increase in cell viability was observed as early as day 1 in cells treated with naïve EVs (p < 0.001) and primed EVs (p < 0.0001) compared to the control group (Figure 2(B)). By days 7 and 14, cell viability was comparable across all groups, confirming the absence of cytotoxic effects.

Figure 2.

Evaluation of target cell viability and osteogenic response following treatment with naïve and primed MSC-derived EVs. (A) Representative live/dead fluorescence images of target cells after exposure to naïve or primed MSC-EVs. (B) quantitative analysis of percent cell viability at days 1, 7, and 14 post-treatment, demonstrating the biocompatibility of both EV types. (C) total DNA content was measured over the time course to assess cell proliferation and survival within the 3D environment. (D) normalized alkaline phosphatase (ALP) activity was quantified at each time point as an indicator of osteogenic differentiation. Scale bars; 100 µm. Data are presented as mean ± standard deviation (n = 3, NS = not significant, ***p < 0.001, ****p < 0.0001).

Total DNA content in encapsulated cells showed a decreasing trend over time, likely due to cellular adaptation from 2D to 3D culture conditions, cell migration from scaffolds, and differentiation-related reductions in proliferation. However, DNA content did not differ significantly between treatment groups at any time point, further supporting the nontoxic nature of the EV treatments (Figure 2(C)).

Assessment of osteoinductive activity via ALP quantification revealed that both naïve and primed MSC-EVs significantly enhanced ALP activity in target cells on day 7 (p < 0.0001 vs. control; Figure 2(D)). Notably, primed MSC-EVs induced significantly higher ALP activity than naïve EVs (p < 0.001), indicating superior osteoinductive potential. This trend persisted through day 14, with both EV groups again showing elevated ALP activity relative to controls (p < 0.0001), and primed EVs maintaining a significantly greater effect.

Collectively, these findings indicate that MSC-derived EVs, particularly those from osteogenically primed cells, enhance osteogenic activity in recipient cells without inducing cytotoxicity. Based on these results, primed MSC-EVs were selected for subsequent investigations.

3.3. Synergistic osteoinductive effects of primed MSC-EVs and a mineralized microenvironment

Following dose standardization, a concentration of 90 µg/mL of primed MSC-derived EVs was selected for further evaluation, based on the enhanced osteoinductive response previously observed at 30 µg/mL. In this phase of the study, osteoinductive EVs were combined with an osteoconductive, mineralized microenvironment (nHAp), and their combined effect on target encapsulated cells was assessed.

Cytotoxicity assays confirmed that neither the primed MSC-EVs nor the mineralized microenvironment induced cytotoxic effects up to 14 days of incubation (Figure 3(A)). Percent cell viability remained statistically unchanged across all treatment groups, indicating biocompatibility (Figure 3(B)). DNA content in encapsulated cells declined over time, likely reflecting cell adaptation from 2D to 3D culture, scaffold egress, and differentiation-driven reductions in proliferation. However, no significant intergroup variation in DNA content was observed, confirming the absence of treatment-related cytotoxicity (Figure 3(C)).

Figure 3.

Target cell viability and osteogenic potential were evaluated following treatment with primed MSC-EVs in a mineralized microenvironment. Representative live/dead fluorescence images of target cells are shown in panel (A), and quantification of cell viability as a percentage is presented in panel (B) after treatment with MSC-EVs and/or nHap. Osteogenic differentiation was assessed by measuring total DNA content (C), normalized alkaline phosphatase activity (D), and calcium deposition (E) at days 1, 7, and 14 post-treatment for each group. Scale bars; 100 µm. Data are presented as mean ± standard deviation (n = 3, NS = not significant, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001).

Significant enhancement of ALP activity was noted as early as day 1 in the EV-only group (p < 0.05 vs. control), suggesting active enzymatic cargo delivery by the primed MSC-EVs. On day 7, normalized ALP levels were significantly elevated in groups treated with EVs alone (p < 0.0001), nHAp alone (p < 0.05), and the combination of EVs and nHAp (p < 0.001), relative to the control (Figure 3(D)). By day 14, the combined treatment group exhibited the highest ALP activity, significantly surpassing the control (p < 0.0001), EV-only (p < 0.0001), and nHAp-only (p < 0.0001) groups. Both individual treatments also remained significantly elevated compared to the control. Mineralization, assessed by calcium deposition, was significantly enhanced in nHAp-treated groups from day 7 onward (p < 0.0001 vs. control and EV group; Figure 3(E)). On day 14, the combined treatment group demonstrated a markedly greater mineralization response than the control (p < 0.0001), EV only (p < 0.0001), and nHAp-only (p < 0.01) groups. Cells treated with nHAp alone also showed significantly increased calcium deposition compared to the control (p < 0.0001). These results underscore the synergistic osteoinductive potential of combining primed MSC-EVs with a mineralized scaffold, as evidenced by significantly enhanced ALP activity and mineral deposition, supporting their promise in bone regenerative applications.

3.4. Cargo study for MSC-EVs

Cargo study of MSC naive and primed EVs: MSC primed EVs had calcium as cargo, whereas naive EVs did not. Both EV types transport ALP; however, primed EVs carried 2.82-fold more than naive EVs. The expression of mRNAs associated with osteogenesis, such as Runx2, was 3.48-fold higher in primed EVs than in naive EVs (Figure 4). The expression of angiogenesis-related mRNAs, such as Vegf-a, was 2.49-fold higher in primed EVs than in naive EVs (Figure 4). Hence, results showed that MSC-primed EVs have osteoinductive and angiogenic potential as cell-free therapies.

Figure 4.

Comparative gene expression study for MSC naïve and primed EVs for osteogenesis and angiogenesis-related genes.

3.5. MSC-EVs and nHap synergize to induce osteogenesis in vivo

Following subcutaneous implantation of hydrogel constructs into C57BL/6 mice and a 4-week in vivo period, histological and mineralization assessments were conducted to evaluate the osteoinductive response of various hydrogel formulations. H & E staining of retrieved tissue sections revealed no evidence of inflammatory cell infiltration or adverse tissue reactions across all experimental groups, indicating excellent biocompatibility of the alginate – PEGDA-based IPN hydrogels, regardless of their biochemical composition (Figure 5). The absence of an inflammatory response is a critical parameter affirming the suitability of these hydrogels as potential biomaterial scaffolds for regenerative applications.

Figure 5.

(A) in vivo evaluation of osteoinductive potential and biocompatibility of IPN hydrogel formulations following subcutaneous implantation in C57BL/6 mice. (B) H & E staining of explanted hydrogels after 4 weeks revealed no inflammatory cell infiltration or adverse tissue response across all groups, confirming the excellent biocompatibility of the alginate–PEGDA-based IPN hydrogels. (C) alizarin red S staining and (D) quantitative analysis of alizarin red S staining, demonstrated minimal mineralization in the scaffold-only control group, moderate calcium deposition in the nHap only and EVs only groups, and a markedly enhanced mineral signal in the nHap + EVs group, indicating a synergistic effect between nano-hydroxyapatite and MSC-derived EVs on in vivo mineralization (n = 3).

Assessment of mineral deposition using Alizarin Red S staining demonstrated distinct differences in calcium accumulation among the groups (Figure 5). The scaffold-only control group showed negligible mineral staining, confirming its non-osteogenic baseline. In contrast, both the nHAp group and the EV only group exhibited positive Alizarin Red staining, indicating moderate mineral deposition likely due to either the osteoconductive properties of nHAp or the osteoinductive signaling of primed MSC-EVs, respectively. Notably, the group containing both nHAp and EVs displayed a substantially stronger and more widespread Alizarin Red signal, reflecting a synergistic enhancement in mineralization. This observation suggests that the combined presence of a mineralized microenvironment and bioactive EVs creates an optimized niche for osteogenic activity.

These findings are consistent with prior in vitro studies suggesting that nHAp enhances the osteoconductivity of hydrogels by mimicking bone mineral content, while MSC-derived EVs contribute to osteoinduction through the delivery of regulatory proteins and RNAs that promote osteoblast differentiation. The pronounced mineralization in the nHAp + EV group supports the hypothesis that the interaction between structural and biological cues plays a crucial role in guiding tissue regeneration. The data reinforce the concept that integrating nano-mineral components with paracrine signals from EVs can significantly boost the functional performance of engineered scaffolds in vivo. Taken together, the results validate the IPN hydrogel platform as a promising strategy for bone tissue engineering and suggest that further exploration in critical-sized defect models is warranted to assess translational potential.

4. Discussion

The present study demonstrates the therapeutic potential of osteogenically primed MSC-EVs in conjunction with a mineralized nHAp-containing hydrogel for bone tissue engineering. A series of in vitro and in vivo experiments were conducted to explore the osteoinductive properties of primed MSC-EVs, their cytocompatibility, and their synergistic effect with a mineralized scaffold, culminating in the evaluation of bone matrix deposition in a subcutaneous implantation model. Initial characterization using FE-SEM revealed no discernible differences in the size or morphology of EVs harvested from naïve and osteogenically primed MSCs, suggesting that the priming process alters EV cargo content without impacting physical characteristics. Indeed, biochemical analyses confirmed that EVs from primed MSCs contained calcium and significantly higher levels of ALP, and mRNAs related to osteogenesis (Runx2), and angiogenesis (Vegf-a) compared to naïve EVs. These findings underscore the impact of osteogenic preconditioning on EV cargo composition and support their enhanced osteoinductive and angiogenic potential. While the cargo analysis of primed MSC-EVs, highlighting increased levels of Runx2, VEGFA mRNA, alkaline phosphatase (ALP), and calcium, is encouraging, the broader biological implications require further investigation. These components are likely to influence osteogenic activity in recipient preosteoblasts by triggering key intracellular signaling networks. For example, Runx2 is widely recognized as a central transcriptional regulator of osteoblast maturation [24]. It may drive the expression of downstream markers such as osteocalcin and type I collagen by engaging the BMP/Smad and MAPK pathways [25]. Likewise, VEGFA is well-known for its role in blood vessel formation, but it also plays a part in bone development by facilitating the coordination between angiogenesis and osteogenesis, primarily through PI3K/AKT and ERK1/2 signaling [25]. The enrichment of ALP and calcium ions may further support matrix mineralization and act as upstream cues for Wnt/β-catenin pathway activation, a signaling route closely associated with commitment to the osteoblast lineage [26]. While the presence of these factors suggests potential mechanisms, dedicated studies are needed to confirm their downstream effects and functional relevance.

Functional internalization studies demonstrated efficient uptake of both naïve and primed EVs by MC3T3 pre-osteoblasts, validating their cellular delivery capability. Importantly, cytotoxicity assays indicated no adverse effects of either EV population on cell viability or DNA content, even after prolonged incubation. On the contrary, an increase in cell viability was observed at early time points, suggesting a potential proliferative or survival-enhancing role of MSC-EVs in target cells.

Notably, EVs from primed MSCs significantly enhanced ALP activity in encapsulated MC3T3 cells compared to both naïve EVs and untreated controls, confirming their superior osteoinductive function. These findings are consistent with previous reports highlighting the paracrine role of EVs in promoting osteogenic differentiation via transfer of functional RNAs and proteins. To further investigate therapeutic relevance, EVs were incorporated into an osteoconductive IPN hydrogel scaffold enriched with nHAp. While both EVs and nHAp individually enhanced ALP activity and calcium deposition in vitro, the combination of primed MSC-EVs and nHAp resulted in the most pronounced osteogenic response. This synergistic enhancement likely arises from the simultaneous provision of biochemical cues (from EVs) and a bone-mimicking microenvironment (via nHAp), thus optimizing the extracellular niche for osteogenesis. In vivo implantation of the hydrogel constructs into a subcutaneous murine model confirmed the biocompatibility of the IPN scaffold, as evidenced by the absence of inflammatory infiltrates in histological sections. Alizarin Red S staining further validated in vitro findings, with the combination group exhibiting the greatest mineral deposition. These results corroborate the hypothesis that osteoconductive scaffolds can be functionally enhanced by pairing with biologically active EVs to mimic the complexity of the native bone regeneration milieu.

Taken together, this study highlights the promise of using primed MSC-EVs as acellular, paracrine modulators of bone regeneration. Their ability to deliver pro-osteogenic and pro-angiogenic cargo without eliciting cytotoxicity, when combined with a structurally supportive and osteoconductive scaffold, presents a compelling strategy for regenerative medicine. Future work should focus on evaluating these constructs in orthotopic, critical-sized bone defect models and assessing long-term outcomes related to vascularization and functional integration.

5. Conclusion

This study establishes the osteoinductive potential of EVs derived from osteogenically primed MSCs and demonstrates their synergistic effect with nHAp-incorporated hydrogels for BTE applications. Primed MSC-EVs exhibited significantly elevated levels of calcium, ALP, and mRNAs (Runx2 and Vegfa), contributing to enhanced osteogenic and angiogenic activity. These vesicles were efficiently internalized by target pre-osteoblasts and showed no cytotoxic effects in vitro. When combined with a mineralized IPN hydrogel scaffold, primed EVs significantly enhanced ALP activity and mineral deposition both in vitro and in vivo. In vivo implantation confirmed biocompatibility and highlighted the synergistic osteogenic effect of combining bioactive EVs with a mineralized microenvironment. Collectively, these findings support the use of primed MSC-EVs as acellular therapeutic agents and underscore the translational potential of EV-integrated biomaterials in bone regeneration strategies. Thus, EVs may be used to build cell-free/secretome-based therapies for bone repair that circumvent the risks associated with current therapies while enhancing safety and efficacy. A 3D system was employed to demonstrate the combined effects of osteoinductive EVs and nHAp to replicate the microenvironment of bone tissue. The uniquely designed, cell-free osteoinductive and osteoconductive scaffold has the potential to stimulate early osteogenesis and accelerate bone regeneration in vivo.

To advance this EV-based, cell-free bone regeneration strategy toward clinical application, future studies will include extended in vivo evaluations (up to 12 weeks) to assess scaffold degradation, bone formation, and long-term biocompatibility. EV release kinetics will be investigated using ELISA and fluorescence tracking, while osteogenic and angiogenic mechanisms will be validated through protein-level assays and tissue staining. Additionally, miRNA sequencing or proteomic profiling of EVs will be performed to identify key bioactive molecules, followed by rescue experiments to confirm the involvement of specific signaling pathways. These steps are essential to establish the safety, efficacy, and translational potential of the platform.

To further support clinical translation, future research should focus on validating this EV-integrated scaffold in orthotopic, critical-sized bone defect models that more closely mimic clinical conditions. Standardizing EV production under GMP-compliant conditions, optimizing long-term storage and delivery strategies, and establishing release kinetics will be essential to meet regulatory and scalability requirements. This work uniquely contributes to the field by demonstrating that osteogenically primed MSC-EVs can be combined with a mineralized, mechanically stable hydrogel scaffold to create a cell-free platform that mimics the native bone environment and enhances early osteogenesis. Unlike traditional grafts or cell-based therapies, this acellular approach offers advantages in terms of safety, storage, handling, and regulatory compliance. The demonstrated biocompatibility, osteoinductive synergy, and in vivo efficacy suggest this system holds strong potential for future development into off-the-shelf, minimally invasive therapeutic solutions for bone defects, non-unions, and reconstructive surgeries, ultimately accelerating the translation of regenerative medicine technologies into clinical practice.

Funding Statement

This manuscript was funded by the Department of Biotechnology (DBT) of India through a Ramalingaswami re-entry fellowship award (BT/RLF/Re-entry/37/2017) and a core research grant (BT/PR41330/MED/32/768/2020).

Article highlights

• The study aimed to enhance bone regeneration by integrating osteogenically primed mesenchymal stem cell-derived extracellular vesicles (MSC-EVs) into a nano-hydroxyapatite (nHAp)-enriched IPN hydrogel scaffold.

• EVs from primed MSCs showed no size/morphology differences from naïve EVs but exhibited higher calcium, ALP activity, and osteogenic/angiogenic mRNAs (Runx2, Vegf-a).

• Both EV types were efficiently internalized by MC3T3 cells with no cytotoxic effects; primed EVs even promoted early cell viability.

• Primed EVs significantly enhanced ALP activity and calcium deposition, especially when combined with the nHAp-containing hydrogel.

• Subcutaneous implantation in mice confirmed scaffold biocompatibility and showed the highest mineral deposition in the primed EV + nHAp group.

• Osteogenic priming enhances EV cargo bioactivity, and their combination with mineralized scaffolds synergistically promotes bone matrix formation.

• This acellular, EV-based platform offers a safe, off-the-shelf alternative to cell-based bone therapies with improved handling and scalability.

• Further studies will assess long-term outcomes, EV release kinetics, and test the system in critical-sized bone defect models under GMP-compliant production.

Author contribution statement

Ketki Holkar designed the study, performed out in vitro and in vivo experiments, collected and interpreted the data, coordinated sample and data acquisition/validation, performed the statistical analyses, prepared the original manuscript. Prasad Pethe, Vaijayanti Kale and Ganesh Ingavle conceptualized and organized this project, monitored the work, contributed to writing (manuscript review and editing), and acquired funding. All authors have contributed to and approved the final version of the manuscript.

Disclosure statement

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Ethical declaration

The authors confirm that all animal experimental procedures were conducted with appropriate ethical approval. The study was approved by the Institutional Animal Ethics Committee (IAEC) of Symbiosis International University, Pune, India [Approval No.: SSBS/IAEC/04–2021].