Despite recent advancements in various in situ bioprinting systems, the simple deposition of bioconstructs can lead to inadequate tissue regeneration and integration with the host tissue23. To address these challenges, stimulus-assisted bioprinting has emerged as a promising approach for modulating the cellular functions of implanted bioconstructs.
To address this issue, we developed a modified bioprinting technique employing an in situ M-field to align magnetically activated particles within a bioink during extrusion. As depicted in the images of Fig. 1a, we fabricated a model of a ‘patient’ with a VML defect in the thigh muscles. The VML defect region was then filled using our newly proposed in situ M-field-assisted bioprinting method, demonstrating that the magnetically activated bioink was stably deposited onto the defective region.
To observe the M-field distribution using the ring magnets, a simulated result showing the cross-sectional M-field distribution for the three ring-shaped magnets (maximum magnetic flux = 0.09 mT) confirmed the formation of a parallel distribution of the applied M-field, aligned with the flow direction of the bioink, shown in the optical image (Fig. 1b). Furthermore, to select the proper concentration of iron oxide nanoparticles laden in GelMa bioink (5 wt%) and the crosslinking condition using UV dose, we simply tested the cell-viability of hASCs of the bioink for various concentrations of the particles and mechanical property for various UV doses. To determine cell-viability of the bioprinted constructs using the M-field (three ring magnets), various concentrations (50 to 500 ng∙mL− 1) of iron oxide nanoparticles (0.23 ± 0.06 µm) were loaded onto 5 wt% GelMA solutions containing hASCs (1×107 cells∙mL− 1) under the UV dose (820 mJ∙cm− 2) and the printing conditions shown in Fig. 1b. The cell viability on 1 and 7 d remained relatively high (~ 90%) from 50 to 200 ng∙mL− 1, but significantly decreased at 500 ng∙mL− 1 (Supplementary Fig. S1a and S1b). These results are similar with previous research, which indicates that the concentrations of iron oxide-based particles around 200 ng∙mL− 1 are generally considered safe, while at higher concentrations, cellular activities including cell viability, growth, and proliferation were markedly reduced19 (Supplemenatary Table S1). Moreover, Supplementary Fig. S2a showed stress-strain curves of the bioconstructs (iron oxide particle concentration = 200 ng∙mL− 1) fabricated using various UV crosslinking conditions (UV dose = 125 to 1650 mJ∙cm− 2). The tensile modulus gradually increased with UV dose. However, similar tensile moduli between bioconstructs fabricated using UV doses over the 820 mJ∙cm− 2 implied that the methacrylate functional groups have been fully crosslinked over the dose. Based on the results, we fixed the iron oxide concentration at 200 ng∙mL− 1 in the bioink and the UV crosslinking condition at 820 mJ∙cm− 2.
The schematic in Fig. 1c illustrates the physical phenomenon affecting magnetizable particles under an M-field, leading to the uniaxial alignment of chained iron oxide particles. This alignment occurs due to the force (Fa) of dipole-dipole interaction and the subsequent magnetically activated torque (τ), where τ = \(\:\stackrel{⃑}{\text{m}}\) × \(\:\stackrel{⃑}{\text{B}}\), with '\(\:\stackrel{⃑}{\text{m}}\)' representing the magnetic moment of the chained iron oxide particles and '\(\:\stackrel{⃑}{\text{B}}\)' the magnetic flux.
For spherical particles under the M-field, the magnetic moment can be calculated as \(\:\stackrel{⃑}{\text{m}}\) = 12.6µoµcr3[(µi-µc)/(µi + 2µc)]Bo, where µ0 is the permeability of vacuum, µc is the relative permeability of the continuous phase (in our case, GelMA hydrogel), µi is the relative permeability of the magnetizable particle (e.g., iron oxide), 'r' is the radius of the spherical particle, and B0 is the magnetic flux24, 25, 26.
Through this physical mechanism, the magnetizable particles can be chained and aligned in the direction of the M-field, as shown in the optical image in Fig. 1c. After the fabrication of the aligned chains of particles in the flow direction, UV light was used to crosslink the GelMA-based bioink simultaneously, maintaining the structural integrity post-extrusion. Finally, the aligned chain particles within the crosslinked GelMA structure served as spatial topographical cues, significantly influencing the alignment and myogenic differentiation of the embedded hASCs. This is crucial for developing biomimetic muscle structures and enhancing the functional properties of the regenerated muscle tissues.
Analysis of magnetic field distribution and its effect on the flowability of the bioink within a nozzle
To determine the M-field distribution generated by the ring magnets, we positioned a varying number of magnets around the nozzle, maintaining fixed geometry for each magnet (inside radius = 6.8 mm, outside radius = 10 mm, thickness = 5.2 mm) (Fig. 2a). The distribution of magnetic flux around the printing nozzle was calculated using the finite element method magnetics (FEMM) software. As expected, the M-field vectors were aligned parallel to the glass nozzle direction (Fig. 2b), facilitating the uniaxial alignment of the iron oxide particles within the bioink.
Furthermore, to observe the effect of the number of magnets on the M-field distribution, simulations were conducted using one-to four-ring magnets. The results shown in Fig. 2b reveal a homogeneous M-field distribution with three-ring magnets. To quantitatively assess the M-field distribution, Fig. 2c and 2d depict the magnetic flux profiles relative to the midplane of the magnet, perpendicular (x-axis), and parallel (y-axis) to the nozzle length. The graphs demonstrate that increasing the number of magnets significantly enhances the intensity and homogeneity of the magnetic flux in both directions. In particular, a homogeneous magnetic flux along the x-axis is important for achieving a uniform distribution of chain-like particles perpendicular to the bioprinted struts. Based on the simulation results, three magnets appear to be sufficient to achieve a homogeneous alignment of magnetic particles within the bioink.
Exposing a magnetically responsive hydrogel to an M-field can alter its rheological properties, including viscosity, by reorganizing magnetic particles within the hydrogel matrix. When a magnetorheological bioink is subjected to an M-field, the magnetized particles tend to align parallel to the M-field lines, which significantly influences the rheological properties of the fluid by aligning the chains of the magnetic particles. Empirically, an externally applied M-field can reduce the wall shear stress of the bioink within the nozzle, thereby improving its flowability.
To observe this effect, we measured the length of the extruded strut after 2 s of extrusion at a fixed pneumatic pressure (125 kPa) (Fig. 2e). The results showed that increasing the number of magnets (and thus the magnetic flux) led to longer extruded bioink lengths, indicating that the flow resistance in the nozzle was significantly reduced owing to the alignment of magnetic particles. Additionally, the volumetric flow rates measured under various pneumatic pressures (50–200 kPa) and different numbers of magnets (0 to 3) indicated higher flow rates with the application of both pneumatic pressure and M-fields (Fig. 2f). These data suggest that external M-fields, aligned with the bioink flow, can significantly lower the wall shear stress in the nozzle, ensuring safer conditions for cells within the bioink than for those not treated with ring magnet.
To evaluate the accuracy of the hypothesis, we extruded iron oxide-incorporated bioinks with and without in situ M-field at various volumetric flow rates (3.1 to 8.2 µL∙s− 1). The cells were stained with live/dead after 3 d and DAPI/phalloidin (Fig. 2g). As expected, cell viability decreased with higher volumetric flow rates in bioinks not exposed to the M-field, indicating that increased flow rates induced cell damage due to high wall shear stress (Fig. 2h). In contrast, bioinks exposed to the M-field showed relatively high cell viability (~ 90%), suggesting a reduction in wall shear stress. Additionally, the cellular alignment shown in Fig. 2i indicates a gradual increase in orientation with the bioink's volumetric flow rate. Notably, bioinks processed using in situ M-field-assisted bioprinting resulted in significantly higher anisotropic organization of cells.
The bioink's flowability was evaluated by comparing two magnet setups: a ring-shaped magnet and a box-shaped magnet, which provide parallel and perpendicular magnetic flux relative to the bioink flow direction as illustrated by the FEMM simulation results, shown in Supplementary Fig. S3a. The optical image in Supplementary Fig. S3b reveals a shorter extruded length with the perpendicular magnet setup, indicating a disturbance in the bioink's flowability. Additionally, volumetric flow rate measurements under various pressures showed a significant reduction with the perpendicular magnet setup (Supplementary Fig. S3c), further demonstrating its negative impact on flowability. These results are further complemented by the rheological evaluation where the vector of the M-field is applied perpendicular to the rotation of the plate (Supplementary Fig. S4a). As a result, the storage modulus (G’) and complex viscosity (η*) of the magnetorheological bioink exhibited a significant increase for the temperature sweep when subjected to a magnetic field of 0.09 mT (Supplementary Fig. S4b-d). This increase can be clearly attributed to the formation of perpendicular iron oxide chains, which impede the rotational flow of the bioink.
Cell alignment and myogenic activities for the cell-constructs bioprinted with various magnetic fields
To assess biological activities, specifically cell alignment due to topographical cues induced by the aligned particle chains, and myogenic activity, we performed staining on the horizontal cross-section of cell-laden bioconstructs composed of 5 wt% GelMA, iron oxide (200 ng∙mL− 1), and hASCs (1 × 107 cells∙mL− 1) fabricated with varying numbers of ring magnets (0 to 3) (Fig. 3a). We performed DAPI and phalloidin staining, orientation mapping (90° was the printing direction), and orientation frequency analysis of hASCs after 7 d to evaluate cell alignment (Fig. 3b).
Quantitative analysis of DAPI/phalloidin images revealed a gradual increase in the cell nucleus aspect ratio, with 1.11-fold (one magnet), 1.39-fold (two magnets), and 1.38-fold (three magnets) increases compared to the no-magnet process (Fig. 3c). Furthermore, we evaluated the alignment of nuclei along the struts using the orientation factor equation, orientation factor = (90° – φ)/90°, where φ is the full width at half maximum (FWHM) of the orientation angle distribution.
Figure 3d shows the distribution of the nucleus orientation factors across five regions perpendicular to the printing direction. For the no-magnet process, cell alignment was predominantly achieved near the wall region, with poor orientation in the central region. However, with the application of magnets, the difference in nucleus orientation between the center and wall regions gradually decreased as the number of magnets increased. Notably, with the three-ring magnets, the orientation in the central region was almost identical to that in the wall region, exhibiting a high orientation factor (approximately 0.9). These results confirmed that bioconstructs processed with M-field assistance exhibited significant and homogeneous cellular alignment, independent of the printed strut region, suggesting more homogeneous myogenesis compared to conventionally bioprinted cell constructs.
To evaluate the myogenic activities of the cells fabricated using the magnet-assisted process, we quantified the myosin heavy chain (MHC) staining images by calculating the positive index (percentage of nuclei expressing MHC), fusion index (percentage of myotubes with two or more nuclei), and maturation rate (percentage of myotubes containing five or more nuclei) (Fig. 3e). While the cells exhibited a high MHC-positive index (approximately 90%) across all groups, the MHC fusion index and maturation rate were accelerated in cells stimulated with a higher number of M-field (Fig. 3f and 3g). These results confirm that the application of the three-ring magnet provides a homogeneous magnetic flux around the printing nozzle, promoting significant cellular alignment and enhancing the myogenic activity of the cells.
Effects of nano- and micro-iron oxide particles laden in the bioink on cellular activities
Figure 4a presents scanning electron microscopy (SEM) images of micro- and nano-sized iron oxide particles, along with optical images showing chained particles within the GelMA matrix printed using M-field-assisted bioprinting (with three ring magnets). The 3D surface mapping images illustrate the diameters of the particle chains in each fabricated structure. The average diameters of the particles were approximately 0.22 µm (nano-sized) and 1.44 µm (micro-sized) (Fig. 4b). Exposure to the in situ M-field resulted in the aggregation of micro-sized iron oxide particles into larger fibers compared to the aggregation of nano-sized particles (Fig. 4c).
Generally, smaller topographical cues, particularly at the nanoscale or submicron scale, provide a more conducive environment for myogenic activities by enhancing cell-matrix interactions, alignment, mechanical signaling, and gene expression owing to a higher surface area-to-volume ratio. Specifically, the myogenic potential of hASCs is highly dependent on micro- and nanoscale topographical cues that affect the dynamic rearrangement of cytoskeletal structures, thereby altering cellular behavior.
To investigate the effect of magnetic chains comprising micro- and nano-sized particles, we incorporated these particles into the hASC-laden GelMA bioink for in situ M-field-assisted bioprinting. Cell proliferation rates, evaluated using the MTT assay, indicated that cells proliferated throughout the culture period in both the micro- and nano-sized iron oxide particle-incorporated bioconstructs (Supplementary Fig. S5a). Figure 4d shows live/dead, DAPI/phalloidin, and DAPI/MHC images of bioconstructs containing micro- or nano-sized iron oxide particles after 1, 7, and 14 d of culture, respectively. The cell viability, as shown in Supplementary Fig. S5b, was relatively high (> 90%), indicating that the incorporation of both types of particles did not adversely affect the cultured cells.
Cellular alignment analysis revealed that bioconstructs containing nano-sized particles exhibited a 1.14-fold higher orientation factor (Fig. 4e) and 1.14-fold higher aspect ratio (Fig. 4f) than those with micro-sized particles. The myogenic activity was quantitatively assessed based on MHC positivity, MHC fusion, and maturation (Fig. 4g). Constructs with nano-sized aligned particles exhibited significantly higher MHC fusion and maturation. Furthermore, the expression levels of myogenic genes (Myod1, Myh1, Myog, and TnT) were significantly higher in the constructs with nano-sized particles than in those with micro-sized particles (Fig. 4h).
These results align with those of previous studies, showing that smaller topographical cues induce more efficient cell alignment and enhance myogenic activity. This phenomenon can be further explained by the coarse topography formed by the larger fibers, which may create irregularities within the GelMA matrix, as depicted in Fig. 4i. Such irregularities can disrupt the continuous and homogeneous cell distribution during bioprinting. In contrast, nano-sized particles tend to form finer fibers, resulting in a more uniform and smaller topography that facilitates better cell alignment along the printed structures. These variations in topography influence cellular behavior and myogenic activities within cell constructs, thereby impacting the overall functionality and integration of bioprinted tissues. Therefore, we posit that aligning nano-sized iron oxide particles through in situ M-field-assisted bioprinting can provide the necessary topographical cues to achieve optimal cellular alignment.
Mechanotransduction effects of a M-field assisted bioprinting process
The presence of an M-field in the bioprinting process can evoke the polarization of iron oxide particles, inducing a magnetic attraction force. In regard to in situ M-field-assisted bioprinting can stimulate mechanotransduction signaling pathways in cells and induce favorable cellular responses. To evaluate this phenomenon, ‘G’ (conventionally bioprinted hASCs-laden GelMA bioconstruct), ‘GIO’ (conventionally bioprinted iron oxide/hASCs-laden GelMA), and ‘GIOM’ (in situ M-field assisted bioprinted hASCs/iron oxide-laden GelMA bioconstruct) are prepared (Fig. 5a).
Mechanotransduction pathways are pivotal in the cellular response to mechanical stimuli and influence diverse biological functions (Fig. 5b). Among these pathways, the hippo signaling pathway, Wnt/β-catenin pathway, and ion channels can play important roles in regulating cellular processes such as cell growth, differentiation, signaling, and homeostasis. The Hippo pathway is sensitive to mechanical signals and governs cell behavior by regulating proliferation, differentiation, and tissue growth27, 28, 29. Alterations in cell shape and cytoskeletal tension induced by mechanical cues modulate Hippo pathway activity, thereby impacting gene expression and cell destiny. Similarly, the Wnt/β-catenin pathway integrates mechanical and biochemical signals to control cell fate, growth, and differentiation during development and tissue maintenance30. Mechanosensitive ion channels, such as stretch-activated channels, contribute to cellular mechanotransduction by converting mechanical forces into electrochemical signals, thereby influencing gene expression, cell motility, and differentiation31. Through an intricate interplay and adjustment, these mechanotransduction pathways collectively steer cellular responses to mechanical stimuli and play key roles in tissue development, maintenance, and pathological processes32. Therefore, we predicted that cells that were mechanically stimulated using in situ M-field bioprinting method could further initiate the aforementioned cell signaling pathways, leading to favorable cellular responses.
To evaluate the accuracy of this hypothesis, cells cultured on the G, GIO, and GIOM bioconstructs were stained using DAPI/phalloidin/piezo1 after 3 d of culture, as shown in Fig. 5a. After extrusion of the magnetically active bioink through an in situ M-field, we noted that piezo1 was expressed significantly more than cells in the G and GIO bioconstructs. As piezo1 is a protein that functions as a mechanosensitive ion channel, meaning that it responds to mechanical stimuli such as tension, pressure, or stretch, we estimated that the developed system can effectively stimulate mechanotransduction activities in laden cells33, 34. Furthermore, using quantitative real-time polymerase chain reaction (qRT-PCR) analysis, we quantified the expression levels of mechanotransduction-related genes, including components of the Wnt signaling pathway (Wnt and β-catenin) (Fig. 5c), hippo signaling pathway (YAP and TAZ) (Fig. 5d), and stretch-activated ion channels (Piezo1 and TRPV2) (Fig. 5e). We observed significant upregulation of these genes in the GIOM bioconstructs. These results suggest that mechanical stimulation induced by in situ M-field bioprinting plays an important role in augmenting cell proliferation and promoting the myogenic differentiation of bioprinted hASCs.
To further assess the impact of the mechanical stimulation provided by the proposed in situ M-field-assisted bioprinting, we compared it with the conventional method of bioprinting using the same bioink containing iron oxide nanoparticles (200 ng∙mL-1), followed by post-printing M-field stimulation, as illustrated in Supplementary Fig. S6a.
Piezo1, a mechanosensitive ion channel important for various physiological processes, including myogenic activity, was used as a marker to evaluate mechanotransduction. We measured piezo1 expression using DAPI and phalloidin staining at 1 d post-fabrication (Supplementary Fig. S6b). The results indicated that piezo1 expression was notably higher in the in situ M-field-assisted bioprinted constructs than in the control group, which received continuous M-field stimulation after printing. This observation was corroborated by the significant upregulation of mechanotransduction-related genes, including Wnt, YAP, and Piezo1 in the in situ M-field-assisted group after 1 d (Supplementary Fig. S6c). Based on the results, the in situ M-field-assisted bioprinting process enhanced early mechanotransduction processes more effectively than post-printing M-field stimulation. This indicates the potential benefits of in situ bioprinting applications, which offer improved cellular responses and promising outcomes for muscle tissue development.
In vitro cellular response to in situ M-field stimulation
To assess various in vitro cellular responses of the G, GIO, and GIOM bioconstructs, we observed live/dead staining at 1 d, DAPI/phalloidin staining at 7 d, DAPI/MHC staining at 14 d, and DAPI/α-actinin staining at 21 d (Fig. 6a). Live/dead images indicated high cell viability (~ 90%) for all three groups, demonstrating that the fabrication printing process was safe (Fig. 6b). DAPI/phalloidin images at 7 d were used to evaluate the cytoskeletal organization of hASCs within the bioconstructs, and the nuclei were quantitatively assessed for aspect ratio and orientation factor (Fig. 6c and 6d). The results showed that cells cultured in the GIOM bioconstruct exhibited a significantly greater aspect ratio, suggesting that these cells continuously received higher mechanical stimulation owing to topographical cues, which affected mechanotransduction pathways. The orientation factor, which can affect to the formation of structurally and functionally coherent muscle fibers, was also comparatively high (~ 0.75 for GIOM, ~ 0.36 and ~ 0.37 for G and GIO, respectively). Additionally, the MTT assay (Fig. 6e) indicated that the cells in all groups proliferated well over time, with cells proliferating at a considerably higher rate (2.1-fold and 1.9-fold at 7 d) compared to G and GIO, respectively, due to the early mechanotransduction effect.
The myogenic capabilities of cells in the bioconstructs were analyzed using immunofluorescence imaging of DAPI/MHC after 14 d and DAPI/α-actinin after 21 d (Fig. 6a). Quantitative analysis of the MHC-positive index, fusion index, and maturation rate revealed that cells in the GIOM group exhibited more efficient development of multinucleated myotubes than those in the G and GIO groups (Fig. 6f).
To further confirm the myogenic activities observed in the immunofluorescence images, the expression levels of myogenesis-related genes (Myog, MHC, and TnT) were quantitatively assessed at 21 d using qRT-PCR (Fig. 6g) and agarose gel electrophoresis (Fig. 6h). Interestingly, the incorporation of iron oxide particles (GIO) into the cell-loaded GelMA bioconstructs did not affect the expression of myogenesis-related genes, whereas exposure to the M-field significantly upregulated the expression of Myog, MHC, and TnT. These results demonstrate the potential of in situ M-field stimulation in the bioprinting process, which enhances various cellular activities and promotes strong myogenic activities.
In vivo works
To further verify the in vitro findings, the bioconstructs (2 × 4 × 1 mm2) were implanted directly into VML defects in mice after printing. As shown in Fig. 7a, approximately 40% of the tibialis anterior (TA) muscles were removed. Subsequently, bioconstructs (G, GIO, and GIOM) were placed at the defect site for implantation (Fig. 7b). Age-matched (SHAM) and non-treated (Defect) mice were selected as positive and negative controls, respectively, to evaluate the regenerative efficacy of these constructs.
Four weeks after implantation, the TA muscles were harvested and subjected to histological evaluation using Hematoxylin and Eosin (H&E) and Masson's trichrome (MT) staining (Fig. 7c). From the images, analyzing parameters including myotube diameter, peripherally nucleated myofibers, and fibrotic areas were quantitatively assessed (Fig. 7d–7f). Compared with mice that underwent defect treatment, those receiving G, GIO, and GIOM bioconstructs exhibited larger muscle fiber diameters, suggesting the facilitation of muscle regeneration through bioconstruct implantation. Notably, the GIOM group displayed the highest muscle diameter, indicating a 1.66-fold increase compared with the defect group. Furthermore, most myofibers in mice that received GIOM were peripherally nucleated, indicating near-complete muscle fiber regeneration. Conversely, substantial fibrotic areas were observed in the defect, G, and GIO groups, indicating significant fibrosis (Fig. 7f). Interestingly, the similarity in the fibrotic areas between the G and GIO groups suggests that the incorporation of iron oxide particles into the bioconstructs did not induce notable fibrosis.
Figure 7g illustrates the TA muscle weights after 4 weeks of treatment, with values for SHAM, defect, G, GIO, and GIOM measured at 42.3, 25.2, 35.7, 36.7, and 39.2 g, respectively. These results indicated a considerable improvement in muscle weight following bioconstruct implantation after VML defects. Additionally, TA muscle functionality was evaluated through grip strength and latency-to-fall assessments at various time points during the 4 weeks implantation period (Fig. 7h and 7i). As shown, grip strength and latency to fall were significantly reduced in all groups immediately after VML defect induction (week 1). However, the mice that received implants demonstrated faster functional recovery than those with the defect alone, highlighting the beneficial effects of the bioconstructs on muscle regeneration. Notably, by the end of the 4 weeks implantation period, the GIOM group exhibited increased grip strength and prolonged latency to fall, suggesting an active role of GIOM in restoring lost functionality in VML defects.
To identify the cellular types and origins within the TA muscle, we used immunochemical staining with markers for the MHC, human leukocyte antigen (HLA), and CD31 to specifically label endothelial cells (Fig. 8a). Assessment of the MHC-positive area (Fig. 8b) revealed a notably higher expression in the GIOM group than in the defect, G, and GIO groups, approaching levels similar to those in the SHAM group. Remarkably, HLA, which has human reactivity, was absent in the SHAM and defect groups but was positively expressed in the implant groups (G, GIO, and GIOM), suggesting the successful integration of implanted hASCs with the host tissue (Fig. 8c). Particularly, a substantial proportion of cells within the TA muscle of the GIOM group exhibited HLA expression, implying that mechanical stimulation facilitated the integration of hASCs with the host tissue.
After a VML defect, the host initiates muscle recovery through the rapid ingrowth of intricate vascular networks. However, as the regeneration process unfolds, these vascular structures undergo remodeling, transitioning into microvessels dispersed among the mature muscle fibers6, 35. Quantification of the vessel area based on CD31 expression, as depicted in Fig. 8d, underscores the significant formation of vascular areas in the VML compared with the initial stages of muscle regeneration. Importantly, our observations revealed the formation of microvessels in mice that underwent implantation; however, the vessel area was notably smaller in the GIOM group.
In general, nonbiodegradable nanoparticles can elicit immune responses by persisting in tissues and interacting with immune cells, potentially leading to chronic inflammation or immune activation36. Their inability to be efficiently metabolized or cleared may prolong their presence in the body and exacerbate immune reactions37,38. Iron oxide nanoparticles generally exhibit low immunogenicity, although their potential to induce immune responses depends on factors such as particle size, concentration, surface characteristics, and the local tissue environment.
To evaluate the immune response of mice that received various bioconstructs, the harvested tissues were stained using MHC II, indicating M1 macrophages, and CD206, indicating M2 macrophages (Fig. 9a). The results indicated that the expression was significantly more pronounced in the defect group and much lower expression of M1 and M2 was observed in the G, GIO, and GIOM groups (Fig. 9b and 9c). In particular, in the GIOM group, the levels of M1 and M2 macrophages were much closer to those in the sham group. While the VML model results alone did not allow us to definitively conclude that the iron oxide nanoparticles used in this study did not significantly affect the immune response, we cautiously inferred that the concentration of these nanoparticles did not influence the immune response in the VML animal model. However, the number of M1 and M2 macrophages varies based on the degree of muscle regeneration.