Advances in 3D bioprinting of immunoregulatory biomaterials for bone regeneration

Abstract

Bone defects caused by trauma, infection, tumour resection, and degenerative diseases remain major clinical challenges because successful regeneration requires coordinated regulation of inflammation, vascularisation, osteogenesis, and bone remodeling. Advances in osteoimmunology have demonstrated that immune responses are integral to each stage of bone healing, shifting biomaterial design from passive biocompatibility towards active and temporally coordinated immunomodulation. Three dimensional bioprinting provides a versatile platform for integrating material, structural, cellular, and biochemical cues within spatially defined constructs, thereby offering new opportunities to regulate the osteoimmune microenvironment. In this review, we first summarise the immune processes and key immune cell populations involved in the inflammatory, proliferative, and remodeling stages of bone repair, and discuss the major bioprinting technologies and bioink design principles relevant to immunoregulatory bone regeneration. We then examine how material composition, structural and surface cues, and biologically active components, including stem cells, cytokines, exosomes, and decellularised extracellular matrix, regulate immune cell states and their interactions with vascular and osteogenic processes. Particular emphasis is placed on moving beyond static macrophage polarisation towards multicellular, stage specific, and spatiotemporally controlled immunoregulation. Finally, current challenges and future directions are discussed, including responsive and multifunctional bioinks, controlled degradation and delivery, mechanistic evaluation, manufacturing standardisation, and clinical translation. Collectively, immunoregulatory bioprinting provides a promising framework for developing bone regenerative constructs that more closely coordinate with the dynamic biology of tissue repair.

Graphical abstract

Highlights

• •Bone regeneration is reviewed from an osteoimmunological perspective.
• •3D bioprinting enables spatially controlled immunoregulatory biomaterial design.
• •Material cues, structural cues, and bioactive cargos regulate immune-osteogenic coupling.
• •Future bioprinted scaffolds should achieve dynamic and stage-specific immune modulation.

1. Introduction

Bone defects caused by trauma, infection, tumour resection, and congenital abnormalities are common in clinical practice. Their treatment aims not only to fill the defect volume but also to restore long term mechanical continuity and recover limb function [[1], [2], [3]]. Bone possesses an intrinsic capacity for self repair and regeneration, which enables spontaneous healing of some small defects. However, when the extent of tissue loss exceeds the regenerative capacity of the local tissue, a critical sized bone defect may develop. Importantly, no universal dimensional threshold defines such defects, as their occurrence is influenced by multiple factors, including anatomical location, species, defect geometry, local vascular supply, and the mechanical environment [4,5]. In these settings, particularly when impaired vascularisation, infection, or mechanical instability is also present, spontaneous healing is often insufficient to restore complete bone architecture and mechanical function, and surgical intervention is therefore frequently required. Bone grafting remains an important clinical approach for the reconstruction of large or complex bone defects. Autologous bone grafts are regarded as the gold standard because they simultaneously provide osteogenic cells, osteoinductive growth factors, and an osteoconductive matrix [6], whereas allogeneic bone grafts are widely used because of their relatively greater availability and regenerative potential [7]. Nevertheless, current grafting strategies remain limited by restricted donor availability or donor site morbidity, infection and reinfection, immune rejection, and variable therapeutic efficacy, which may result in delayed healing, nonunion, or repair failure in some patients [8,9]. Therefore, developing new regenerative strategies capable of improving the stability and reproducibility of bone regeneration within complex local microenvironments remains of considerable clinical importance.

Bioprinting provides an important technological foundation for constructing regenerative scaffolds with spatial controllability and multidimensional regulatory capacity [23]. Compared with conventional fabrication methods, bioprinting enables the spatial integration of multiple material components within a single scaffold, allowing precise regulation of pore architecture, compositional distribution, and local mechanical properties [[24], [25], [26]]. Composite bioinks constructed from printable hydrogels, such as gelatin methacryloyl (GelMA), alginate, and hyaluronic acid methacrylate (HAMA), together with inorganic phases, including calcium phosphate (CaP), hydroxyapatite (HA), β-tricalcium phosphate (β-TCP), and bioactive glass, can balance printability with the establishment of an osteogenic microenvironment. These bioinks can also influence early host immune recognition through ion release, surface functional groups, and mineralised interfaces [[27], [28], [29], [30]]. It should be noted that the immune response following material implantation has long been regarded as an adverse reaction that should be avoided. Excessive or persistent inflammation can induce tissue damage and fibrotic responses, leading to the formation of a peri-implant fibrous capsule that restricts vascular and cellular infiltration and ultimately compromises the quality of osseointegration [27,31,32]. Based on this understanding, early material design emphasized bioinertness, aiming to attenuate the intensity of immune responses by reducing cell–implant interactions [33]. Although this strategy can, to some extent, mitigate acute inflammation, bone defect regeneration is highly dependent on cell recruitment, vascularisation, and tissue remodeling. Therefore, simply suppressing immune responses may also weaken the capacity of biomaterials to actively regulate regeneration.

Studies in osteoimmunology have shown that immune responses do not merely exert detrimental effects during bone repair, but instead constitute an indispensable biological component throughout the initiation, progression, and remodeling phases of bone regeneration [34]. In the early stage after injury, the recruitment of innate immune cells, including neutrophils and monocytes/macrophages, together with the release of inflammatory mediators, contributes to the clearance of necrotic tissue and provides initial signals for repair activation [35]. Subsequently, inflammatory resolution is accompanied by remodeling of the immune cytokine profile and is closely coupled with angiogenesis, mesenchymal stem cell (MSC) recruitment, and osteogenic differentiation. During the remodeling phase, adaptive immunity and bone metabolic regulation jointly influence bone mass, structural maturation, and long-term homeostasis [36]. Macrophages play a central role in osteoimmunology. An early M1-like phenotype contributes to antimicrobial defense and debris clearance, whereas excessive or prolonged M1 macrophage responses can inhibit osteogenesis and promote fibrosis. By contrast, the M2-like phenotype is associated with inflammatory resolution, pro-angiogenic activity, and matrix remodeling, thereby facilitating the establishment of a microenvironment that supports osteogenesis [37]. Accordingly, the design philosophy of bone-regenerative biomaterials has gradually shifted from passively avoiding immune activation to actively modulating immune responses, with an emphasis on controllable guidance of immune intensity and temporal dynamics to improve osseointegration quality and the reproducibility of bone repair.

In this conceptual transition, bioprinting provides a practical platform for coordinating immunomodulation with osteogenesis [38]. On the one hand, ions such as silicon (Si) and calcium (Ca) released from printable bioactive glass systems can regulate the local microenvironment and modulate the macrophage cytokine profile, thereby promoting a reparative transition while enhancing endothelial cell-associated angiogenic signaling [39]. Magnesium-containing printed scaffolds or systems with controlled magnesium ion release have also been shown to suppress pro-inflammatory signaling and promote M2-like macrophage polarisation, which in turn indirectly enhances osteogenesis-related gene expression and mineralisation [40]. On the other hand, the precise control of structural parameters enabled by bioprinting makes the spatial coupling of immune cell infiltration and cytokine gradients a designable variable [41]. Meanwhile, pore size, interconnectivity, and curvature features can influence the depth and uniformity of immune cell infiltration, thereby shaping the duration of inflammation and local immune resolution. Macroscale mechanical matching and interfacial micromotion control may further reduce the risk of chronic immune activation induced by wear debris and stress concentration [42]. Bioprinting also facilitates the localized delivery and spatiotemporal distribution of immunoregulatory cargoes within scaffolds, such as anti-inflammatory small molecules, immunomodulatory nucleic acids, or mesenchymal stem cell (MSC)-derived exosomes, allowing inflammatory resolution and osteogenic signaling to be regulated in closer accordance with the temporal sequence of bone repair [43,44]. Based on these advantages, this review focuses on the design rationale and strategic framework of bioprinting-mediated immuno-osteogenic regulation. It systematically summarizes the immune chronology of bone repair and the roles of key immune cells, and further discusses recent advances and challenges in immunomodulatory printed scaffolds from three dimensions: material composition, surface morphology, and immunoactive substances(Table 1). Collectively, this review aims to provide a reference for the translational application of bioprinting strategies in bone defect regeneration (Fig. 1).

Table 1.

Comparative evaluation of representative immunomodulatory strategies for 3D-printed and bioprinted bone regeneration.

Representative strategy	Immunoregulatory mechanism and controllability	Safety and major limitations	Cost and manufacturing considerations	Clinical maturity	bone-defect context	Representative reference
Sr-CSH/GelMA bioink with BMSCs	Sustained Sr-, Ca- and Si-related ion release modulates macrophage responses and supports osteogenesis. Regulation can be adjusted through inorganic-phase content and degradation kinetics.	Biological effects remain concentration- and time-dependent. Living-cell incorporation increases requirements for viability and quality control.	Material preparation is relatively straightforward, although cell expansion and sterile manufacturing increase complexity.	Demonstrated in rat critical-sized calvarial defects; evidence remains preclinical.	Critical-sized craniofacial defects and defects with impaired osteogenic activity. Additional reinforcement is required for load-bearing sites.	[10]
bdECM-MA/Si-CaP bioink with BMSCs	Combines tissue-specific matrix cues with Si-CaP-derived ionic signals and promotes a transition from early inflammation to a reparative immune state.	Requires control of residual dECM immunogenicity, ceramic dosage and photopolymerisation conditions. Contributions of individual components are difficult to separate.	Decellularisation, chemical modification, ceramic preparation and cell loading create considerable manufacturing complexity.	Sequential immunoregulation and bone repair have been validated preclinically.	Large or poorly healing defects requiring coordinated immune regulation, vascularisation and osteogenesis.	[11]
LUT@ZIF-8/GelMA bioink with BMSCs	ZIF-8 enables sustained Zn2+ and luteolin delivery, integrating immunoregulation, antioxidation and antibacterial activity.	Therapeutic effects depend on MOF degradation and Zn2+ exposure. Long-term biosafety and individual contributions of Zn2+ and luteolin require clarification.	MOF synthesis and drug loading increase formulation complexity but remain more standardized than cell-derived biologics.	Validated in preclinical cranial defect models.	Inflammatory or infection-prone defects. Direct evidence in established infected bone defects remains limited.	[12]
DLP-printed HA scaffolds with optimized pore size	Precisely defined pore architecture regulates macrophage behavior, tissue infiltration and vascularisation without exogenous biologics.	The optimal pore size is material- and model-dependent and cannot be generalized across scaffold systems.	Avoids costly biologic cargos, but requires optimized ceramic printing and sintering.	Additively manufactured HA is relatively mature, whereas geometry-directed osteoimmunomodulation remains preclinical.	Craniofacial and large aseptic defects requiring vascular and tissue ingrowth.	[13]
Negative-Gaussian-curvature β-TCP scaffold	Negative curvature modulates macrophage signaling and enhances BMP-2/VEGF associated vascularized osteogenesis. Geometric parameters are highly designable.	Curvature is coupled with pore architecture, permeability and mechanical properties. Ceramic brittleness remains a limitation.	Requires high-precision scaffold design and ceramic manufacturing but no exogenous biologics.	Supported by a critical-sized rabbit radial defect model.	Large segmental defects, particularly where vascularized bone ingrowth is required.	14]
CaP-coated MEW-PCL scaffold	Ordered microfibres and CaP-derived signals jointly regulate macrophage mechanotransduction and osteogenic responses.	Slow PCL degradation and limited bulk mechanical strength restrict stand-alone application in large defects.	PCL and CaP are accessible materials, although MEW is relatively slow for large constructs.	Osteoimmune effects have been validated preclinically.	Bone–implant interfaces and composite bone scaffolds. Additional reinforcement is required for major load-bearing defects.	[15]
Bioprinted vascular channels with BMSCs and endothelial cells	Continuous channels spatially organise BMSCs and endothelial cells and enhance angiogenesis–osteogenesis coupling.	Multicellular constructs require strict control of cell source, viability and channel patency. Direct immune mechanisms remain insufficiently defined.	Multicellular culture and sterile printing substantially increase manufacturing burden.	Demonstrated in rat critical-sized calvarial defects.	Particularly relevant to ischemic and poorly vascularized defects.	[16]
IFN-γ/Laponite-containing bioink	Sequential signaling preserves early inflammatory activation and subsequently promotes a reparative transition through Laponite-derived ionic cues.	Excessive or prolonged IFN-γ exposure may sustain inflammation, making release kinetics critical.	Recombinant cytokine use increases cost, whereas Laponite is relatively inexpensive and improves printability.	Demonstrated in preclinical bioprinted vascularized bone constructs.	Large or poorly vascularized defects requiring temporal coordination of inflammation and angiogenesis.	[17]
IL-4-loaded GO/BP-functionalised scaffold	Local IL-4 delivery regulates macrophages, while GO and BP provide additional pro-angiogenic and osteogenic cues.	Continuous IL-4 exposure may interfere with necessary early inflammation. Long-term nanomaterial degradation also requires evaluation.	Cytokine loading and nanomaterial functionalisation increase cost but avoid cell culture and vesicle purification.	Bone regeneration and angiogenesis have been demonstrated in animal models.	Aseptic inflammatory and vascularisation-deficient defects. Less suitable for uncontrolled active infection.	[18]
VEGF/BMP-2 delivery in HA/PCL scaffold	Spatially sustained delivery of VEGF and BMP-2 coordinates angiogenesis and osteogenesis within a mechanically supportive scaffold.	Both factors have dose-dependent adverse effects, including abnormal angiogenesis and ectopic bone formation.	Recombinant growth factors substantially increase manufacturing and quality-control costs.	Supported by a rabbit tibial large weight-bearing defect model.	Particularly relevant to large weight-bearing and ischemic segmental defects.	[19]
Schwann-cell-derived exosomes in GelMA/SilMA with BMSCs	Exosomes provide multi-target regulation of osteogenesis, angiogenesis and neural regeneration, while SilMA prolongs local retention.	Exosome composition and potency depend on parental-cell state. The additional use of BMSCs further increases biological variability.	Exosome production, purification and cell expansion impose substantial manufacturing demands.	Neurovascularized bone regeneration has been demonstrated in rat cranial defects.	Craniofacial and neurovascular-compromised defects, particularly where vascular regeneration is limited.	[20]
USC-Exos/GelMA/DBM bioink	Combines prolonged exosome release with bone-specific matrix cues to regulate inflammation, angiogenesis and osteogenesis.	Standardization is complicated by variability in both exosomes and DBM. The individual effects of the two components remain difficult to distinguish.	Requires exosome production, DBM processing and multi-component quality control.	Demonstrated in preclinical alveolar or mandibular bone-defect models.	Alveolar and complex craniofacial defects requiring simultaneous immune and vascular regulation.	[21]
Dental-follicle-derived dECM/GelMA bioink	Tissue-specific dECM attenuates excessive inflammatory signaling and provides matrix cues for periodontal and alveolar regeneration.	Immunological effects vary with tissue source and processing. Residual DNA, sterilisation and batch variability remain major concerns.	Decellularisation and biochemical quality control increase manufacturing requirements but avoid recombinant biologics.	Supported by a beagle periodontal defect model.	Particularly suitable for periodontal, alveolar and bone–ligament interface defects.	[22]

Fig. 1.

Immunoregulatory mechanisms of bioprinted biomaterials for bone regeneration. Bioprinted scaffolds regulate innate and adaptive immune responses through material composition, structural cues, and bioactive signals, thereby coordinating inflammation, angiogenesis, osteogenesis, and bone remodeling to promote bone regeneration.

2. Immune processes and key immune cells in bone repair

Bone repair is a typical immune-mediated regenerative process, and its outcome depends on the dynamic balance of post-injury immune responses in terms of intensity, temporal progression, and spatial distribution [45]. As shown in Fig. 2, the early stage of bone injury is dominated by inflammation, during which hematoma formation, activation of the coagulation cascade, and the release of damage-associated molecular patterns (DAMPs) collectively trigger innate immune responses [46]. Neutrophils and monocytes/macrophages rapidly infiltrate the injured site. Among them, M1 macrophages participate in pathogen and necrotic tissue clearance through inflammatory mediators such as tumor necrosis factor-α (TNF-α) and interleukin-1β (IL-1β), while also regulating osteoclast activation. As repair progresses, the local immune microenvironment gradually shifts from an inflammatory state toward a pro-reparative state. M2 macrophages secrete factors such as transforming growth factor-β (TGF-β) and vascular endothelial growth factor (VEGF), thereby promoting mesenchymal stem cell (MSC) recruitment, angiogenesis, and osteogenic differentiation, and facilitating the transition from the inflammatory phase to the proliferative phase [34]. In the subsequent remodeling stage, osteoblasts and osteoclasts maintain the dynamic coupling between bone formation and bone resorption through signaling pathways such as receptor activator of nuclear factor-κB (RANK), receptor activator of nuclear factor-κB ligand (RANKL), and osteoprotegerin (OPG). Meanwhile, osteocytes participate in mineral maturation, bone mass maintenance, and mechanically adaptive remodeling through molecules such as dentin matrix protein 1 (DMP1) and insulin-like growth factor-1 (IGF-1) [47]. Based on this immune-temporal framework, this section summarizes the key immune events occurring at different stages of bone repair and further focuses on the functional roles and potential regulatory targets of neutrophils, macrophages, and adaptive immune cells in immuno-osteogenesis, thereby providing a biological basis for the design and evaluation of subsequent immunomodulatory bioprinting strategies.

Fig. 2.

Schematic illustration of immune-mediated bone regeneration during the inflammatory, proliferative, and remodeling stages. Macrophage polarisation regulates the coupling between osteoclasts, osteoblasts, and osteocytes through inflammatory cytokines and osteogenic factors, thereby coordinating bone resorption, angiogenesis, osteogenesis, and bone remodeling.

2.1. Inflammatory phase

After bone injury, local vascular disruption leads to blood extravasation and the formation of a fracture hematoma. The coagulation cascade is rapidly activated, and the fibrin network, together with platelet adhesion and aggregation, establishes a temporary scaffold within the defect region [48]. During the first few days, the hematoma becomes progressively organized. Its fibrin matrix provides an initial three-dimensional support for the migration, adhesion, and proliferation of immune cells and mesenchymal progenitor cells, while also serving as the primary site for early cytokine signaling and cell–matrix interactions within the defect area [49]. Meanwhile, platelet-derived factors, such as transforming growth factor-β (TGF-β) and platelet-derived growth factor (PDGF), participate in early chemotactic and reparative signaling and are associated with subsequent mesenchymal stem cell (MSC) recruitment and callus formation [50]. Under ischemic and hypoxic conditions, necrotic cells and damaged extracellular matrix (ECM) release multiple types of damage-associated molecular patterns (DAMPs), which activate local innate immune responses through pattern-recognition receptor (PRR)-mediated signaling axes, including Toll-like receptors (TLRs). These signals can also interact with and amplify the complement system, forming key upstream events that initiate and propagate inflammation [51]. Complement cleavage products C3a and C5a not only participate in immune cell recruitment and inflammatory mediator production, but also interact with the process of bone regeneration. Animal studies have shown that deficiency of C3 or C5 alters fracture repair, suggesting that complement exerts functional effects during the inflammatory phase that influence subsequent bone regeneration [52].

Neutrophils are usually the first immune cells to arrive at the defect site within hours after injury. They mediate antimicrobial defense and debris clearance, and further amplify inflammatory signaling by releasing proteases through degranulation, such as elastase, producing reactive oxygen species (ROS), and secreting various chemokines and cytokines. These processes promote monocyte recruitment and shape the early immune microenvironment [35,53]. Neutrophils are not only effector cells responsible for clearing necrotic material and cellular debris. Within the fracture hematoma, they can also synthesize and deposit specific matrix components, forming an early emergency ECM that may facilitate subsequent matrix remodeling and progenitor-cell entry [54]. In addition, neutrophil extracellular traps (NETs) play roles in infection control and signal amplification. However, excessive NET formation or delayed NET clearance can increase tissue cytotoxicity and prolong the duration of inflammation, and is therefore considered to be associated with chronic inflammation and unfavorable repair tendencies [55,56]. Thus, the key issue during the inflammatory phase is not to suppress neutrophil responses, but to limit excessive inflammatory activation and promote their timely transition toward a resolvable tissue-repair program after the completion of early defense and clearance functions.

Following neutrophil infiltration, circulating monocytes continuously enter the injury site and differentiate into macrophages. These macrophages contribute to inflammatory resolution and microenvironmental reprogramming by phagocytosing necrotic debris and apoptotic neutrophils, a process known as efferocytosis, making them central regulators of the transition from the inflammatory phase to the reparative phase [57]. Changes in the macrophage cytokine profile during this stage are closely associated with the initiation of angiogenesis, MSC recruitment, and the establishment of osteogenesis-related signaling. Multi-omics studies have shown that the spatiotemporal interactions among immune, stromal, and mesenchymal cells during early bone repair significantly influence subsequent callus quality and repair efficiency [36]. Overall, the intensity and duration of inflammation are decisive for subsequent repair. Moderate, controllable, and resolvable inflammation facilitates MSC homing, angiogenic factor expression, and osteogenic initiation. By contrast, excessive or prolonged inflammation can increase cell apoptosis, induce protease- and ROS-mediated matrix damage, disrupt metabolic homeostasis, and potentially promote fibrotic tendencies and unfavorable healing through persistent danger signaling and sustained activation of complement and PRR pathways [58]. In the context of trauma accompanied by systemic inflammation or severe concomitant injuries, DAMP-mediated immune imbalance has been associated with impaired fracture healing. Interventions targeting high-mobility group box 1 (HMGB1) have also shown potential to improve healing in animal models, further indicating that the inflammatory phase represents a therapeutically targetable window [59].

From the perspective of immunomodulatory material design, the objective during the inflammatory phase is not to eliminate inflammation as early as possible, but rather to preserve the transient inflammatory response required for host defence, clearance of necrotic tissue, and initiation of regenerative signalling while limiting excessive ROS production, protease activity, NET formation, and persistent DAMP–PRR signalling, and simultaneously promoting efferocytosis and timely resolution of inflammation. Accordingly, the early post implantation period requires precise control over the magnitude and duration of immune stimulation rather than sustained delivery of potent anti inflammatory signals. For ion releasing or degradable materials, this implies avoiding excessive early burst release of ions, particulates, or degradation products, particularly for functional ions with relatively narrow therapeutic windows. At the same time, sufficient structural integrity should be maintained during this phase to prevent overly rapid degradation from generating additional damage associated signals or inducing mechanical instability.

2.2. Proliferative phase

As pathogen burden decreases and necrotic tissue is progressively cleared, the inflammatory response enters a resolution phase. The local cytokine profile is remodeled from a pro-inflammatory-dominant state toward a pro-reparative state, characterized by a decline in pro-inflammatory mediators and an increase in pro-resolving or pro-reparative signals, such as interleukin-10 (IL-10), transforming growth factor-β (TGF-β), interleukin-4 (IL-4), and interleukin-13 (IL-13). This transition is accompanied by a shift in the chemotactic network from sustained immune-cell recruitment toward stromal-cell homing and tissue reconstruction [60]. At this stage, functional remodeling of macrophages represents a central event in inflammatory resolution. On the one hand, macrophages promote the resolution of inflammation by engulfing apoptotic neutrophils, a process known as efferocytosis, and thereby induce a shift in their own secretory profile toward a pro-reparative phenotype. On the other hand, macrophage-derived pro-angiogenic and tissue-remodeling factors, such as vascular endothelial growth factor (VEGF), provide essential paracrine support for revascularization within the defect region, creating favourable conditions for the migration, survival, and differentiation of osteoblasts and mesenchymal stem cells (MSCs) [34,61]. Therefore, immune resolution during the reparative phase is not merely an anti-inflammatory process, but rather a macrophage-mediated tissue-reconstruction program. Its temporal window and intensity directly affect regenerative efficiency and callus quality.

Bone regeneration during the reparative phase is highly dependent on effective vascularisation. Under the regulation of local growth factors and immune mediators, vascular endothelial cells undergo migration, proliferation, and lumen formation, thereby establishing a nascent vascular network and improving oxygen and nutrient supply within the defect region. At the same time, endothelial cells support osteogenic lineage progression through paracrine signaling [62]. Studies have shown that bone tissue contains a specific pro-osteogenic vascular subtype closely associated with osteogenic activity, namely CD31ˆhi Type H vessels. These vessels are located in close proximity to bone progenitor-cell niches and can mediate angiogenesis–osteogenesis coupling through signaling pathways such as Notch, thereby influencing the rate and quality of bone formation [63,64]. Regarding growth factor-mediated angiogenesis, VEGF not only promotes vascular invasion, but also influences osteogenic commitment by regulating endothelial phenotypes. Notably, the effect of VEGF on angiogenesis–osteogenesis coupling is highly dose-dependent. Either insufficient or excessive VEGF levels may reduce the efficiency of vascularisation-mediated osteogenesis [65,66]. These findings indicate that vascularisation during the reparative phase should not be assessed solely by vessel quantity. Instead, greater attention should be paid to vascular phenotypes and signaling states that are functionally coupled with osteogenesis.

Alongside vascularisation, MSCs and osteoprogenitor cells gradually enter the defect region and initiate osteogenic lineage differentiation. Chemotactic signals retained from the inflammatory phase or generated during the reparative phase, such as the C-X-C motif chemokine ligand 12/stromal cell-derived factor-1 (CXCL12/SDF-1)-related axis, are thought to participate in MSC recruitment and homing. Meanwhile, remodeling of the immune cytokine profile provides a more favourable microenvironment for MSC osteogenic differentiation and the maintenance of osteoblast function [67]. Subsequently, osteoblasts secrete bone matrix and gradually form woven bone. Depending on the anatomical site and defect conditions, intramembranous ossification and endochondral ossification may occur with different relative contributions or proceed in parallel. Overall, the key feature of the reparative phase is the coordinated progression of immune resolution and osteogenic differentiation. A pro-reparative immune microenvironment creates conditions for sustained bone formation by regulating angiogenic factors, pro-osteogenic endothelial phenotypes, and matrix remodeling. Conversely, insufficient inflammatory resolution or restricted vascularisation may lead to inadequate regenerative signaling, disrupted matrix deposition and mineralisation rhythms, reduced woven-bone quality, and ultimately impaired remodeling potential and mechanical recovery [37,68]. Vascularisation and osteogenesis are not always strictly synchronized across different bone-healing models. Recent intravital imaging studies suggest that, in certain bone-regeneration contexts, transient temporal dissociation may occur between vascular invasion and osteogenic repair, highlighting the importance of precise spatiotemporal regulation during the reparative phase of bone healing [69].

As bone repair progresses into the proliferative phase, the primary goal of immunomodulation gradually shifts from restraining excessive inflammation to promoting inflammatory resolution, angiogenesis, MSC recruitment, and the establishment of immuno osteogenic signalling. At this stage, material design should minimise persistent stimulation of early pro inflammatory responses while progressively providing cues that support reparative macrophage functions, vascularisation, and osteogenic differentiation. In contrast to the inflammatory phase, in which controlling the peak magnitude of release is particularly important, the proliferative phase is better suited to moderate, sustained, or sequential delivery strategies. These approaches can align the temporal availability of biological signals such as IL 4, IL 10, SDF 1, and VEGF, together with osteoactive ions such as Si, Ca, and Sr, with vascular ingrowth and the formation of new bone matrix. Meanwhile, the progressive degradation of hydrogels or composite scaffolds should gradually create additional space for cell migration and tissue ingrowth. However, scaffold mass loss should not substantially outpace the formation of new tissue, as premature degradation may compromise structural integrity and load transfer before sufficient tissue regeneration has occurred.

2.3. Remodeling phase

On the basis of woven bone and hard callus formed during the reparative phase, bone tissue enters a remodeling stage that lasts for several months or even longer. Histologically, this stage is characterized by the gradual replacement of woven bone with lamellar bone, which has a more organized structure and higher degree of mineralisation. Trabecular orientation, cortical thickness, and medullary cavity morphology are progressively optimized according to loading demands, eventually approaching the pre-injury anatomical and mechanical state [70]. This stage is not simply an increase in bone mass. Instead, it is centred on coupled renewal between bone resorption and bone formation. Through the cyclic activity of basic multicellular units (BMUs) on specific bone surfaces, following an activation, resorption, and formation (ARF) sequence, old bone is removed, microcracks are repaired, and new bone is deposited, thereby achieving microstructural optimisation and mechanical adaptation [71]. During BMU activity, osteoclasts resorb bone and form resorption lacunae. Subsequently, osteoblasts deposit osteoid at the same site, followed by progressive mineralisation and the formation of lamellar bone within discrete bone units. This process constitutes the fundamental kinetic framework of bone remodeling [72].

Bone metabolic coupling during the remodeling phase is jointly regulated by multiple bone-resident cells. Among them, osteocytes act as central sensors of mechanical stimuli and integrators of signaling. They can regulate osteoclastogenesis and osteogenic activity by modulating the receptor activator of nuclear factor-κB ligand/osteoprotegerin (RANKL/OPG) axis, Wnt signaling inhibitors such as sclerostin, and various paracrine factors, thereby translating changes in mechanical loading into regulatory signals for bone remodeling [73]. The RANKL/receptor activator of nuclear factor-κB (RANK)/OPG pathway is a pivotal signaling axis governing osteoclastogenesis and the intensity of bone resorption. RANKL promotes osteoclast precursor differentiation and survival, whereas OPG functions as a decoy receptor that neutralizes RANKL and inhibits osteoclastogenesis. This axis plays a decisive role in both physiological bone remodeling and various forms of pathological bone loss [74,75]. In addition to this classical pathway, multiple coupling factors mediate bidirectional regulation between osteoclasts and osteoblasts, allowing bone resorption sites to transition into bone formation programs while maintaining spatial consistency during renewal. These coupling mechanisms are considered an important biological basis for the conversion of woven bone into lamellar bone and the completion of cortical bone reconstruction during the remodeling phase.

Unlike the early stages, which are dominated by innate immunity, the remodeling phase involves a markedly increased contribution from the bone marrow immune microenvironment. Interactions among adaptive immune cells, osteocytes, osteoblasts, and osteoclasts become more prominent, forming a typical osteoimmunological regulatory network [76]. T cells can directly promote osteoclastogenesis by expressing RANKL, while the cytokine profiles of different T-cell subsets can alter the balance of osteoclast–osteoblast coupling. T helper 17 (Th17)-associated interleukin-17 (IL-17) can upregulate RANKL and macrophage colony-stimulating factor (M-CSF) expression in osteoblasts and mesenchymal cells, thereby enhancing osteoclastogenesis and bone resorption. Under chronic inflammatory conditions, this process is associated with bone loss [77]. By contrast, immunoregulatory factors produced by regulatory T cells (Tregs), such as interleukin-10 (IL-10), are thought to suppress excessive osteoclastogenesis and promote the restoration of inflammatory homeostasis, thereby indirectly supporting coupled balance during remodeling [78,79]. B cells are also important components of the bone marrow niche. They can secrete OPG and participate in the inhibition of osteoclastogenesis. However, under specific inflammatory or age-related conditions, B cells may also contribute to RANKL signaling, altering the OPG/RANKL balance and thereby influencing the direction of bone remodeling [80]. Therefore, bone mass restoration and structural maturation during the remodeling phase largely depend on adaptive immune-mediated regulation of the RANKL/OPG axis and its long-term dynamic balance with local inflammatory status.

The sensitivity of the remodeling phase to immune homeostasis makes it vulnerable to disruption by persistent low-grade inflammation, chronic infection risk, or immune abnormalities. When the local environment is exposed to sustained pro-inflammatory factors, such as tumor necrosis factor-α (TNF-α), interleukin-1β (IL-1β), and interleukin-6 (IL-6), or when T cell-derived pro-osteoclastogenic signaling is enhanced, osteoclastogenesis and bone resorption may remain persistently elevated. This can cause bone resorption to exceed bone formation, resulting in insufficient bone mass, limited trabecular maturation, and incomplete cortical reconstruction, ultimately compromising long-term stability and mechanical recovery [[81], [82], [83]]. In the context of persistent infection or inflammation, the remodeling phase may also show both slowed bone repair and reduced structural quality of the regenerated bone, including insufficient bone formation and delayed mineralisation or histological maturation. These changes ultimately lead to long-term structural instability and an increased risk of reinjury [84]. Therefore, the quality of bone repair depends not only on controlling the intensity of inflammation during the inflammatory phase and ensuring efficient bone formation during the reparative phase, but also on maintaining sustainable homeostatic renewal through immune–bone metabolic coupling during the remodeling phase over a longer time scale.

As repair progresses into the remodeling phase, the focus of immunomodulation shifts further from inflammatory resolution and new bone formation towards the maintenance of immune homeostasis and long term coupling between osteoblasts and osteoclasts. In particular, persistent low grade inflammation, dysregulated RANKL signalling, and excessive osteoclastic activity should be avoided. Accordingly, this stage is not well suited to continued strong anti inflammatory stimulation or sustained macrophage directed polarisation. Instead, exogenous immunomodulatory cues should gradually diminish, while low level material derived signals that support matrix mineralisation, tissue maturation, and balanced bone metabolism are maintained. For degradable scaffolds, the degradation rate should progressively become coordinated with new bone formation, lamellar bone maturation, and the redistribution of physiological loads. Excessively rapid degradation may result in premature loss of structural support and local accumulation of degradation products, whereas overly slow degradation may prolong the foreign body response, impede tissue replacement, or interfere with normal mechanically adaptive remodeling. Therefore, the terminal degradation of an ideal scaffold should not be dictated solely by a predefined timeframe, but should instead be synchronized as closely as possible with the progression of bone formation and remodeling.

2.4. Immune cells in bone repair

Bone repair is not governed by a single immune cell population or a single inflammatory phenotype, but instead depends on the temporally coordinated recruitment, functional transitions, and interactions of innate and adaptive immune cells. Neutrophils primarily participate in pathogen clearance and initiation of inflammation during the early stage of injury. Macrophages contribute throughout the repair process, including inflammatory resolution, angiogenesis, initiation of osteogenesis, and bone remodeling. Dendritic cells bridge innate and adaptive immunity through antigen presentation and regulation of T cell responses, whereas T and B cells further modulate inflammatory homeostasis, osteogenesis, and osteoclastogenesis through cytokines and signalling pathways such as the RANKL/OPG axis [85,86]. Recent single cell and multi omics studies have further revealed substantial functional heterogeneity within these immune cell populations, with their roles varying according to the type of injury, stage of repair, and local microenvironment. Accordingly, osteoimmunomodulation cannot be adequately explained by a simplified model of M1 versus M2 macrophage polarisation [36,86]. From the perspective of biomaterial design for bone regeneration, distinguishing the stage specific functions of different immune cell populations provides an important biological basis for moving beyond single macrophage centred regulation towards multicellular and temporally coordinated immunomodulation.

2.4.1. Macrophages and bone resident macrophages

Macrophages are among the most important cellular mediators linking immune regulation with tissue regeneration after bone injury. They comprise both macrophages recruited from circulating monocytes and tissue resident macrophages that persist within bone. During the early stage of injury, macrophages contribute to host defence and tissue clearance by phagocytosing necrotic cells and tissue debris and by secreting inflammatory mediators such as TNF-α, IL-1β, and IL-6. They can also influence MSC recruitment and the initiation of osteogenesis through signalling pathways involving cyclooxygenase-2 (COX-2)/prostaglandin E2 (PGE2) and oncostatin M (OSM). As repair progresses, macrophages remove apoptotic neutrophils through efferocytosis and gradually adopt secretory programmes increasingly characterised by pro-resolving and tissue reparative mediators, including IL-10, TGF-β, and VEGF, thereby coordinating inflammatory resolution, angiogenesis, and osteogenic differentiation [36,87]. Although the conventional M1/M2 classification is useful for describing broad functional tendencies of macrophages, it does not fully capture their in vivo states. Recent multi omics studies of bone repair have identified sequentially emerging pro-inflammatory, reparative, and anti-inflammatory macrophage populations, indicating that macrophages undergo a continuum of functional transitions during bone healing rather than occupying two stable and mutually exclusive terminal phenotypes [36]. Therefore, ideal immunomodulation during bone regeneration should not aim to sustain or continuously enhance M2 polarisation. Instead, it should preserve an appropriate early inflammatory response while promoting a timely transition towards inflammation resolution and tissue reconstruction.

Within the macrophage compartment, bone resident macrophages, also termed osteal macrophages or osteomacs, exhibit distinct tissue localisation and functions compared with inflammatory macrophages derived from circulating monocytes. Osteomacs are distributed along endosteal and periosteal surfaces and are frequently found in close association with mature osteoblasts, where they form canopy-like structures over bone forming surfaces. Functional studies have shown that depletion of CD169+ osteomacs markedly reduces osteoblast numbers and impairs both intramembranous and endochondral ossification, indicating that these cells can directly support osteoblast maintenance and bone formation independently of osteoclasts [88]. More recent evidence further demonstrates that osteomacs engulf apoptotic osteoblasts and acquire a distinct transcriptional state during efferocytosis, accompanied by increased expression of factors such as VEGFA that promote vascular formation and the accumulation of osteogenic progenitor cells [89]. Osteomacs should therefore not be regarded merely as conventional inflammatory effector cells, but rather as integral components of the osteogenic microenvironment at bone surfaces. In the design of immunomodulatory materials for bone regeneration, future strategies should therefore extend beyond regulating the M1/M2 balance of recruited macrophages and consider whether the material can preserve or reconstruct a bone surface niche that supports osteomac survival, efferocytosis, and osteoblast supportive functions.

2.4.2. Neutrophils

Neutrophils are among the earliest immune cells to infiltrate bone defects in large numbers after injury. In response to DAMPs, chemokines, and complement signals, neutrophils are rapidly recruited to the fracture hematoma, where they contribute to pathogen control and clearance of necrotic tissue through phagocytosis, degranulation, ROS production, and the formation of neutrophil extracellular traps (NETs). They also release chemotactic mediators that facilitate the subsequent recruitment of monocytes and other immune cells [85,90]. Neutrophils therefore function not merely as transient amplifiers of inflammation, but as important regulators of the early immune microenvironment following bone injury. Emerging evidence further suggests that neutrophils may directly contribute to the initiation of regeneration. Cai et al. showed that neutrophils exhibiting an N2 related phenotype under appropriate interleukin-8 (IL-8) stimulation could secrete SDF-1α and promote BMSC recruitment through the SDF-1/CXCR4 axis and downstream PI3K/AKT and β-catenin signalling, thereby facilitating the initiation of bone regeneration [91]. The effects of NETs on bone related cells also appear to be highly context dependent. Some studies suggest that NET associated IL-17A can promote osteogenic differentiation of MSCs, whereas more recent evidence indicates that NETs may suppress osteoclastogenesis under specific conditions [92].

Nevertheless, neutrophils also exert pronounced context dependent and potentially detrimental effects. Excessive ROS production, protease release, and persistent NET formation can damage the extracellular matrix, amplify inflammatory responses, and delay inflammatory resolution, ultimately impairing osteogenesis. The recently proposed N1/N2 classification of neutrophils should therefore be interpreted with caution, as direct evidence supporting clearly defined N1 and N2 phenotypes during bone repair remains substantially more limited than that available for macrophage functional states [90]. Accordingly, a more appropriate goal for neutrophil directed osteoimmunomodulation is not simply to increase the proportion of N2 neutrophils, but to regulate the magnitude, activation state, and duration of neutrophil responses. Such temporal control should allow neutrophils to fulfil their early defensive and clearance functions while enabling timely resolution of their highly inflammatory state and supporting subsequent cell recruitment and tissue regeneration.

2.4.3. Dendritic cells

Dendritic cells (DCs) are professional antigen presenting cells that provide a critical link between innate and adaptive immunity. They can recognise damage associated or material associated antigens and regulate T cell activation and differentiation through antigen presentation and costimulatory signalling. Following bone injury, DC abundance may increase during the early inflammatory phase, and DCs can influence CD4⁺ and CD8⁺ T cell responses through cytokine networks, thereby indirectly modulating the subsequent magnitude of inflammation and the state of bone metabolism [86]. In addition, DCs and osteoclasts both arise from the myeloid haematopoietic lineage and can exhibit functional overlap under specific microenvironmental conditions. In chronic inflammatory settings, DCs may enhance osteoclastogenesis by promoting Th17/IL-17 responses and RANKL related signalling, whereas certain factors derived from DCs can exert opposing effects. Chen et al. demonstrated that IFN-λ1 secreted by activated DCs inhibited osteoclast formation by suppressing NF-κB/NFATc1 signalling and NLRP3 inflammasome activation, thereby attenuating inflammatory bone destruction [93]. These findings indicate that the effects of DCs on bone metabolism are determined by their activation state and their interactions with T cells and osteoclasts, rather than by a simple classification as proinflammatory or anti-inflammatory cells.

Compared with macrophages, the direct involvement of DCs in physiological fracture repair remains relatively understudied, with much of the available evidence derived from models of inflammatory bone loss, infectious bone disease, and osteoimmune disorders. Recent multicellular coculture and single cell studies have begun to suggest that DCs may influence bone regeneration by modulating interactions among osteoblasts, osteoclasts, and T cells, although the relevant DC subsets, temporal windows, and specific functions remain insufficiently defined [94,95]. DCs therefore represent a markedly underappreciated cellular target in osteoimmunomodulatory biomaterial research. For bioprinted scaffolds intended for long term implantation, future studies should assess whether material composition, surface chemistry, and degradation products alter DC maturation, antigen presentation, and subsequent T cell responses, rather than focusing exclusively on macrophage surface markers.

2.4.4. T cells

T cells are among the most important adaptive immune cells involved in bone repair, and different T cell subsets can exert markedly distinct or even opposing osteoimmune effects. CD4⁺ T cells can further differentiate into Th1, Th2, Th17, and regulatory T cells (Tregs). interferon-γ (IFN-γ) produced by Th1 cells can suppress osteogenic signalling and sustain a proinflammatory microenvironment when present at persistently high levels, whereas the effects of IL-17 produced by Th17 cells are strongly context dependent. In the setting of acute bone repair, IL-17 can enhance osteogenesis related gene expression in MSCs, and IL-17A derived from γδ T cells has also been shown to promote bone formation in certain models. By contrast, under chronic inflammatory conditions, sustained IL-17 signalling can enhance osteoclastogenesis by increasing RANKL expression [85,96]. CD8⁺ T cells also exert complex effects, with early clinical and animal studies suggesting that an increased abundance of terminally differentiated CD8⁺ T cells is associated with impaired fracture healing.

Recent studies of Tregs have substantially expanded our understanding of their functions during bone repair. Tregs have traditionally been considered to suppress excessive inflammation and osteoclast formation through mediators such as IL-10 and TGF-β, but emerging evidence indicates that their functions extend beyond immunosuppression. Wu et al. found that Tregs within the fracture callus exhibited high amphiregulin expression and directly promoted the proliferation and differentiation of osteoprogenitor cells through PI3K/AKT signalling, whereas transient depletion of Tregs markedly increased the risk of nonunion [97]. Subsequent studies identified a reparative CCR8⁺ Treg subset at sites of bone injury. The CCL1/CCR8 axis promoted recruitment of these cells to the injured tissue, where they directly enhanced skeletal stem cell accumulation and osteogenic differentiation through the secretion of progranulin [96]. These findings demonstrate that adaptive immune cells can promote bone repair not only indirectly through suppression of inflammation, but also directly by providing tissue regenerative signals to osteoprogenitor populations. Tregs and other T cell subsets with tissue reparative functions may therefore represent important targets for the development of next generation immunomodulatory biomaterials.

2.4.5. B cells

B cells primarily develop within the bone marrow, where they share a common microenvironment with osteoblasts, stromal cells, and osteoclast precursors and can therefore directly participate in the regulation of bone metabolism. B cell abundance changes dynamically during fracture healing. They can enter the fracture site during the early stage of injury, decline before the end of the inflammatory phase, and subsequently increase again during hard callus formation and remodeling [86]. Their effects on bone metabolism are largely mediated through regulation of the RANKL/OPG balance and cytokine secretion. Certain B cell populations can express RANKL and promote osteoclastogenesis, whereas B cells and plasma cells also represent important sources of OPG within the bone marrow and can restrain excessive bone resorption by neutralising RANKL. Thus, B cell function cannot be categorised as uniformly osteogenic or osteoresorptive, but instead depends on the composition of B cell subsets and the local RANKL/OPG ratio [83]. In addition, B cells can modulate osteoblasts, T cells, and other immune populations through mediators such as TNF-α, CCL3, IL-6, and IL-10, thereby contributing to immune homeostasis and bone remodeling within the bone marrow.

Regulatory B cells (Bregs) may represent an important B cell subset involved in bone repair. Clinical studies have shown that IL-10⁺ B cells increase during the early stage of normal fracture healing, whereas this early increase is absent in patients with delayed healing or nonunion, suggesting that impaired Breg function may be associated with persistent inflammation and poor bone repair [98,99]. Bregs may also help maintain a regenerative immune environment indirectly by promoting Treg formation and suppressing T cell derived inflammatory signals such as TNF-α and IFN-γ. However, compared with macrophages and Tregs, direct studies targeting specific B cell subsets in biomaterials for bone regeneration remain limited, and part of the available mechanistic evidence is derived from relatively early clinical investigations. Accordingly, for immunomodulatory bioprinted materials, B cells should currently be regarded as an important immune component that warrants inclusion in future evaluation frameworks, rather than as a sufficiently validated target for the design of materials directed towards any single B cell subset.

3. In vivo immune responses to implants

After implantation, the implant surface immediately adsorbs plasma proteins, complement components, and DAMPs, forming the earliest protein corona. This adsorbed layer is recognized by neutrophils, monocytes/macrophages, and dendritic cells through integrin receptors and pattern-recognition receptors (PRRs), thereby initiating a local innate immune response [100]. Subsequently, neutrophils are the first to participate in acute inflammatory clearance, whereas monocytes are recruited to the implant interface and differentiate into macrophages, which become the central immune cells determining the long-term fate of the implant [101]. Early M1 macrophages contribute to pathogen clearance and amplification of damage signals by secreting pro-inflammatory cytokines, including TNF-α, IL-1β, and interleukin-6 (IL-6). However, if the M1 response persists, it may induce foreign body giant cell formation, fibroblast activation, and fibrous encapsulation, ultimately leading to a foreign body reaction and implant failure. By contrast, timely transition toward M2 macrophages can promote inflammatory resolution, angiogenesis, mesenchymal stem cell (MSC) recruitment, and osteogenic differentiation through factors such as interleukin-4 (IL-4), TGF-β, VEGF, and bone morphogenetic protein-2 (BMP-2), thereby facilitating tissue integration and osseointegration (Fig. 3) [102]. The immune process is regulated by multiple implant-related factors, including material composition, surface topography, pore architecture, wettability, mechanical compatibility, and degradation products. For example, 90 nm TiO₂ honeycomb-like nanostructures can promote macrophage pseudopodia formation and activate RhoA/ROCK-related signaling, biasing macrophages toward an M2 phenotype and subsequently enhancing BMP-2 release and osseointegration [102]. In addition, highly porous biodegradable zinc–lithium (Zn-Li) scaffolds promote an increased M2/M1 ratio through the synergistic effects of Zn/Li ion release, surface micro/nanotopography, and a gyroid macrostructure, and show stronger alkaline phosphatase (ALP) and osteocalcin (OCN) expression as well as enhanced new bone ingrowth in vivo [103]. Therefore, the immune response to implants should not be regarded simply as an adverse inflammatory reaction, but rather as a dynamic transition from early inflammation toward immunomodulation and tissue regeneration. The key lies in suppressing persistent M1-dominated inflammation while promoting the formation of an M2-like immune microenvironment, thereby enabling stable integration between the implant and host tissue.

Fig. 3.

Implant-mediated immune responses and normal tissue-healing processes jointly determine the fate of implants. IFN-γ, interferon-γ; IL, interleukin; TNF, tumor necrosis factor [100]. Copyright 2020, Nature Portfolio.

4. Bioprinting technologies

Bioprinting technologies determine how biomaterials, cells, and bioactive components are spatially organised within three dimensional constructs and therefore represent a critical technological link between the design of immunomodulatory materials and the realisation of scaffold function. Different printing modalities vary substantially in their fabrication principles, compatible material ranges, printing resolution, cell loading capacity, manufacturing speed, and ability to spatially organise multiple materials, resulting in distinct suitability for immunomodulatory bone repair scaffolds [29,104]. Extrusion based bioprinting can process hydrogels and composite bioinks across a relatively broad viscosity range and readily incorporate cells, inorganic particles, and bioactive factors, making it well suited for constructing multimaterial and gradient scaffolds. Inkjet bioprinting offers high deposition speeds and precise droplet placement but is constrained by bioink viscosity and cell concentration. Digital light processing enables rapid layerwise photopolymerisation of photosensitive bioinks through projection based exposure and provides advantages in fabrication speed and geometric fidelity. However, its application relies on photocrosslinkable material systems and requires careful consideration of the effects of photoinitiators, light dose, and optical scattering on cells and bioactive components. Melt electrowriting can generate highly ordered microscale fibrous architectures and is particularly useful for improving scaffold mechanical performance and creating microenvironments that guide cellular alignment. However, because the melt processing conditions generally preclude direct encapsulation of living cells, it is more appropriately used as a structural reinforcement framework for cell laden hydrogels. Volumetric bioprinting, in contrast, uses three dimensional light fields to solidify entire constructs within a short period, thereby substantially reducing the cumulative cellular stress associated with prolonged layer by layer fabrication. Nevertheless, its broader application remains limited by material compatibility, optical transparency requirements, and challenges in manufacturing at large tissue scales [104,105]. Accordingly, no single bioprinting technology is universally optimal. The choice of printing modality should instead be matched to the spatial resolution, cell loading strategy, bioactive factor stability, and mechanical requirements of the intended immunomodulatory approach. For bone regeneration, future efforts should focus not only on improving geometric fidelity, but also on exploiting different printing modalities to spatially compartmentalise and temporally organise immunomodulatory signals within engineered constructs.

4.1. Extrusion-based bioprinting

Extrusion-based bioprinting is one of the most widely used bioprinting modalities in tissue engineering. Its basic principle involves pneumatically, piston-, or screw-driven extrusion of bioinks through a nozzle to generate continuous and stable filaments, which are deposited layer by layer along predefined paths to construct three-dimensional structures [104,106]. Compared with inkjet and light-based bioprinting, extrusion-based bioprinting accommodates a broader range of rheological properties and can process GelMA, alginate, collagen, hyaluronic acid, decellularised extracellular matrix (dECM), and composite hydrogels containing inorganic particles or nanomaterials. It can also support relatively high cell densities and incorporate bioactive components such as proteins, growth factors, and exosomes [29,106]. Multi-nozzle systems, coaxial extrusion, and microfluidic-assisted extrusion further enable spatial compartmentalisation of different materials and cell populations, providing a technical basis for engineering bone constructs with compositional gradients, cellular heterogeneity, and region-specific biological signals [106]. Importantly, extrusion-based bioprinting should be distinguished from conventional thermoplastic extrusion. The former generally processes cell-laden or bioactive bioinks under cytocompatible conditions, whereas melt extrusion of thermoplastic polymers such as polycaprolactone (PCL) requires elevated processing temperatures and therefore usually precludes direct encapsulation of living cells during extrusion. Nevertheless, thermoplastic extrusion can be combined with hydrogel bioprinting to generate hybrid constructs with enhanced mechanical support.

For immunomodulatory bone regeneration, a major advantage of extrusion-based bioprinting is its ability to integrate material-derived immunomodulatory cues, living cells, and locally delivered bioactive components within a single three-dimensional construct, while further controlling their spatial distribution through printing paths and multimaterial patterning. Yu et al. developed a BMSC-laden GelMA/Sr-CSH composite bioink in which Sr-CSH nanowires not only improved the printability and mechanical properties of GelMA, but also modulated macrophage phenotypes and improved the inflammatory microenvironment through the release of bioactive ions, ultimately resulting in pronounced bone regeneration in a rat critical-sized calvarial defect model [10]. The significance of extrusion-based bioprinting therefore extends beyond the simple encapsulation of cells within three-dimensional hydrogels. It also enables spatial integration of immunomodulatory materials with osteogenic cell sources, allowing material-mediated immune regulation to act synergistically with cell-mediated bone formation. Extrusion-based printing also offers substantial compatibility with multifunctional components. Dutta et al. incorporated polyphenol-derived carbon quantum dots into a GelMA bioink and used single-nozzle extrusion to fabricate constructs containing human bone marrow mesenchymal stem cells (hBMSCs) and endothelial cells. The resulting system exhibited favourable shear-thinning and structural recovery properties, promoted macrophage transition towards a repair-associated phenotype, and enhanced angiogenic and osteogenic responses. In addition, the construct enabled near-infrared-responsive drug release, thereby integrating immunomodulatory, antitumour, and bone regenerative functions within a single printed system [107]. For pathological scenarios such as post-tumour-resection bone defects and infection-associated defects, extrusion-based bioprinting is particularly suitable for integrating cells, nanomaterials, ions, and drug delivery modules within bioinks to establish local microenvironments with multiple coordinated biological functions.

However, the major limitations of extrusion-based bioprinting arise from the inherent trade-offs among printing fidelity, structural stability, and cell protection. Increasing bioink viscosity or extrusion pressure, or reducing nozzle diameter, generally improves filament stability and geometric fidelity, but simultaneously increases the shear stress experienced by cells, which may compromise cell viability and subsequent function. Conversely, reducing viscosity can mitigate extrusion-induced cellular damage but often results in structural collapse and poor shape fidelity after printing [3,106]. This trade-off is particularly relevant to bone tissue engineering, where scaffolds must provide not only a cytocompatible environment but also sufficient macroscopic structural stability. To address this challenge, Yang et al. developed coaxially extruded hollow hydrogel scaffolds and subsequently achieved rapid cell loading through controlled structural compression and recovery after printing. This approach avoided direct exposure of cells to the extrusion process while improving cell loading efficiency within large constructs [3]. Such strategies also demonstrate that simultaneous extrusion of cells and bioinks is not the only viable approach. Post-printing cell loading, embedded printing, and multimaterial hybrid fabrication may all provide effective means of mitigating the trade-off between printing performance and cellular function.

4.2. Inkjet bioprinting

Inkjet bioprinting is a droplet-based biofabrication technique that typically uses thermal, piezoelectric, or other drop-on-demand actuation mechanisms to deposit low-viscosity liquids containing cells or bioactive components as discrete microdroplets at predefined locations. Unlike extrusion-based bioprinting, which relies on continuous filament deposition to construct the main scaffold architecture, inkjet bioprinting offers high deposition speed and precise spatial positioning, making it particularly suitable for the localised, discrete, and low-dose patterning of different cell populations or soluble signals [108,109]. Accordingly, in immunomodulatory bone regeneration, the principal value of inkjet bioprinting may lie less in fabricating the bulk scaffold itself and more in exploiting droplet-level deposition to control the spatial distribution of immunomodulatory cues across different regions. By varying local cell density, cell type, or bioactive factor concentration, spatially distinct immune microenvironments can be established within a single construct, providing a means to reproduce the heterogeneous inflammatory and regenerative signals present in bone defects. More recently, the integration of microfluidic systems with droplet-based printheads has further improved cell manipulation and deposition precision, enabling selective positioning of different cell populations and the fabrication of heterogeneous tissue constructs [109]. Gao et al. used thermal inkjet printing to co-deposit human bone marrow-derived MSCs with hydroxyapatite or bioactive glass nanoparticles within poly(ethylene glycol) dimethacrylate (PEGDMA) hydrogels, achieving three-dimensional spatial localisation of both cells and osteoactive inorganic components while maintaining high cell viability and osteogenic differentiation capacity [110]. Although this study did not directly assess immunomodulatory effects, it demonstrated that inkjet printing can locally co-deposit cells and functional components without relying on continuous scaffold formation. For immunomodulatory bone repair, this capability may be more appropriately exploited for regional patterning of cytokines, chemokines, exosomes, or distinct immune-related cell populations rather than simply reproducing the material-loading strategies used in extrusion-based printing. In other words, the potential value of inkjet bioprinting lies primarily in enhancing the spatial selectivity of immune signals rather than increasing the overall material-loading capacity of the scaffold.

This technological role is, however, constrained by the limited range of compatible materials. Stable droplet generation generally requires low viscosity and relatively low solid content, whereas high cell densities, high inorganic particle contents, or highly viscous polymer systems can lead to unstable droplet formation, nozzle clogging, and non-uniform cell deposition [109,111]. This limitation conflicts with several common requirements in bone tissue engineering, where large-volume scaffolds often demand relatively high material concentrations, sufficient mechanical strength, and substantial inorganic mineral content. Consequently, the application range of inkjet bioprinting in bone regeneration is comparatively narrow, and the technique is better suited as a precision deposition module for cells and bioactive signals in combination with prefabricated scaffolds, photocrosslinked hydrogels, or other printing modalities. At present, relatively few studies have directly used inkjet bioprinting to regulate the immune microenvironment of bone repair. This suggests that its advantages in this field remain centred largely on the potential for spatially controlled delivery, rather than on a mature application framework comparable with that of extrusion-based bioprinting.

4.3. Digital light processing bioprinting

Digital light processing (DLP) bioprinting constructs three-dimensional structures by projecting predefined two-dimensional light patterns onto photosensitive biomaterials, thereby inducing simultaneous photopolymerisation across an entire layer before subsequent layers are sequentially cured. Compared with extrusion-based bioprinting, which relies on line-by-line deposition, a major advantage of DLP is that fabrication time is relatively insensitive to the geometric complexity of each layer, while high structural resolution and good geometric reproducibility can also be achieved [105,111]. These characteristics give DLP a distinct technological role in immunomodulatory bone regeneration. Its principal advantage does not lie in accommodating a broader range of materials, but rather in its ability to precisely define pore size, pore geometry, curvature, connectivity, and local microarchitecture, thereby influencing the spatial distribution, morphology, and material-contact environment of immune cells after scaffold infiltration. DLP is therefore particularly suitable for investigating and constructing immunomodulatory systems in which immune regulation is mediated by geometric architecture itself, rather than relying exclusively on material composition or exogenous bioactive factors.

This advantage has received relatively direct experimental support in osteoimmunomodulation studies. Xiong et al. used DLP to fabricate HA scaffolds with identical material composition but pore sizes of 400, 600, and 800 μm, thereby isolating a single structural variable to compare the immune and regenerative effects of different pore architectures. The 600 μm pore size more favourably regulated macrophage phenotypes and was associated with changes in IFN-β- and HIF-1α-related signalling, together with improved angiogenesis and bone formation [13]. By exploiting the high degree of structural control provided by DLP, this study minimised confounding between material composition and geometric effects and thereby established a clearer relationship among structural parameters, immune responses, and bone regeneration. It should be noted, however, that the study used acellular HA bioceramic scaffolds and therefore, strictly speaking, represents DLP-based additive manufacturing rather than conventional cell-laden bioprinting. Nevertheless, from the perspective of immunomodulatory material design, it provides important evidence that high-precision light-based fabrication can convert scaffold architecture itself into a controllable immunomodulatory variable. The further development of DLP towards true bioprinting systems requires this structural precision to be integrated with cytocompatible and bioactive material platforms. Song et al. used a photosensitive GelMA/HAp composite ink to fabricate high-precision porous scaffolds and optimized GelMA degree of substitution, polymer concentration, and HAp content to balance photocuring performance with osteoactivity. The resulting constructs achieved a printing resolution of approximately 100 μm and supported favourable osteoblast adhesion, proliferation, and bone formation [27].

This developmental direction is, however, constrained by the requirements of photosensitive material systems. Because DLP relies on photopolymerisation, materials must possess appropriate photoreactivity, optical transparency, and crosslinking kinetics. High cell densities, elevated inorganic particle contents, or optically opaque components can increase light absorption and scattering, causing the actual fabrication fidelity to fall below the theoretical resolution [105]. In addition, photoinitiator concentration, irradiation wavelength, and light dose must be carefully balanced to achieve sufficient crosslinking while preserving cellular function. This trade-off is particularly relevant to bone tissue engineering because increasing the content of HA, bioceramics, or other mineral phases can improve osteoactivity and mechanical performance while simultaneously impairing light propagation and curing uniformity. DLP is therefore particularly suitable for bone repair systems that require high structural precision, complex pore architectures, and finely controlled spatial microenvironments, whereas its compatibility is generally more limited for highly viscous, particle-rich, densely cell-laden, or highly multimaterial bioinks. Compared with extrusion-based bioprinting, DLP offers clear advantages in structural fidelity and manufacturing efficiency, but is more constrained by material selection and photocuring requirements. Accordingly, the two technologies occupy distinct application niches in the development of immunomodulatory bone scaffolds.

4.4. Melt electrowriting

Melt electrowriting (MEW) combines melt extrusion with electrostatic stretching and stabilisation of polymer jets, enabling highly ordered fibres, typically at the micrometre scale, to be deposited along predefined trajectories. This provides substantially greater structural control over fibre orientation, spacing, pore size, and multilayer stacking than conventional melt extrusion [112,113]. Unlike extrusion-based bioprinting, which primarily relies on hydrogels to carry cells and bioactive components, the principal value of MEW lies in the precise engineering of cell-scale physical microenvironments. Ordered fibrous networks can modulate cell adhesion, spreading, and migration, while fibre orientation and local pore architecture can influence cytoskeletal organisation and mechanotransduction. MEW is therefore particularly well suited for investigating how structural cues regulate immune cell behaviour. In bone tissue engineering, such high-precision microfibrous architectures can also regulate pathways for tissue ingrowth and vascular invasion while maintaining high pore interconnectivity, thereby providing a structural basis for geometry-mediated immunomodulation [112,114].

The structural advantages of MEW have gradually extended beyond the promotion of osteogenesis to the regulation of osteoimmune responses. Daghrery et al. demonstrated that altering fibre orientation and spacing in MEW-fabricated PCL scaffolds markedly affected macrophage behaviour. Ordered and aligned fibres promoted a more elongated macrophage morphology and increased repair-associated signals such as IL-10, whereas randomly organised fibres elicited more pronounced proinflammatory characteristics. Larger fibre spacing also favoured osteogenic differentiation, indicating that MEW can simultaneously influence immune status and tissue-specific regeneration through microarchitectural parameters [115]. Building on this structural control, MEW can be further combined with surface functionalisation. Shi et al. modified highly ordered MEW-PCL scaffolds with a CaP coating and found that changes in surface morphology and Ca²⁺ release promoted macrophage transition towards a repair-associated phenotype through PI3K/AKT- and cAMP-PKA-related signalling, respectively, thereby further enhancing BMSC osteogenesis and in vivo bone formation [15]. Thus, the value of MEW in immunomodulatory bone scaffolds extends beyond improved structural resolution. It enables the physical microenvironment to be precisely defined first, after which chemical cues can be superimposed through mineral coatings or other surface functionalisation strategies, allowing structural regulation to act synergistically with material-mediated osteoimmunomodulation.

Nevertheless, the application boundaries of MEW are relatively well defined. Because the process requires polymer melting, currently compatible materials are dominated by thermoplastics such as PCL, and living cells, exosomes, and most protein-based immunomodulatory factors cannot be directly processed together with the polymer during printing [112,113]. In addition, fabrication efficiency, attainable construct height, and material diversity remain constrained for large-scale three-dimensional structures, while polymer microfibres alone are generally insufficient to provide both the mechanical support required for load-bearing bone defects and the complex biological activity needed for regeneration [116]. MEW is therefore better positioned as a high-precision structural framework within immunomodulatory bone constructs, where it can be combined with cell-laden hydrogels, bioactive coatings, or other bioprinting modalities, rather than replacing extrusion- or light-based bioprinting as a complete cell-delivery platform. This integration of a precisely engineered structural framework with biological functionality may represent the most practically relevant direction for the future development of MEW in osteoimmunomodulation.

4.5. Volumetric bioprinting

Volumetric bioprinting (VBP) differs fundamentally from conventional point-by-point, line-by-line, or layer-by-layer fabrication. It is typically based on tomographic reconstruction principles, in which a series of two-dimensional light patterns are projected from multiple angles into a rotating vat containing a photosensitive bioresin, allowing the target region to accumulate a sufficient light dose for simultaneous solidification throughout the entire three-dimensional volume [117,118]. This layer-free fabrication strategy enables centimetre-scale cell-laden constructs to be produced within seconds to tens of seconds and reduces the dependence of complex overhanging geometries on supporting materials [117]. The development of VBP in bone tissue engineering has progressively evolved from rapid structural fabrication towards the reconstruction of increasingly complex bone-like microenvironments. Bernal et al. demonstrated early that centimetre-scale cell-laden constructs could be generated within tens of seconds using volumetric bioprinting, including structures displaying trabecular bone-like geometries and angiogenic sprouting, thereby showing that this approach could combine architectural complexity with high cell viability [117]. Building on this concept, Gehlen et al. incorporated cocultures of MSCs and endothelial cells into volumetrically printed constructs and generated heterogeneous bone-like tissues within soft GelMA bioresins, extending VBP from simple cell encapsulation towards the coordinated reconstruction of osteogenic and vascular microenvironments [119]. This progression is particularly relevant to immunomodulatory bone regeneration because bone repair is not governed by osteogenic cells alone, but depends on sustained interactions among immune, vascular, and stromal cell populations. The rapid whole-volume fabrication characteristic of VBP therefore provides a technical basis for the future spatial organisation of multiple cell populations within three-dimensional constructs. As tissue-specific requirements have become more demanding, research has increasingly shifted towards bioresin chemistry and the recreation of mineralised bone-like niches. Duquesne et al. compared different photocrosslinking mechanisms for osteoid-like matrix fabrication and showed that a step-growth gelatin system based on norbornene-thiol chemistry could improve printing resolution while reducing photoinitiator concentration and light exposure, while still supporting osteogenic differentiation of dental pulp-derived stem cells. This work further advanced VBP from rapid fabrication towards biomimetic recreation of the osteoid niche [120]. More recently, de Wildt et al. combined polymer-induced liquid-precursor mineralisation with VBP and mitigated the optical scattering caused by mineral precursors, enabling rapid fabrication of bone-like constructs containing high cell densities and amorphous mineral precursors. Cell viability exceeded 90%, and MSC differentiation towards an early osteocyte-like phenotype was observed [121]. Collectively, these advances indicate that the key developmental trajectory of VBP in bone tissue engineering is shifting from maximising fabrication speed towards simultaneously addressing the compatibility of high cell density, mineralisation, and complex bone-like microenvironments.

Direct evidence supporting VBP for osteoimmunomodulation, however, remains very limited. Existing bone-related studies have primarily focused on MSC osteogenesis, vascularisation, osteoid-like microenvironments, and mineralisation, whereas strategies that use VBP to actively organise macrophages, neutrophils, T cells, or temporally regulated immune mediators have not yet developed into a mature research framework. VBP should therefore currently be regarded as a promising platform for constructing complex multicellular osteoimmune models and future immunomodulatory scaffolds, rather than as a technology with already established advantages in osteoimmunomodulation. In addition, VBP is highly dependent on optical conditions. High cell densities, mineral particles, and other optically opaque components increase light scattering and reduce fabrication accuracy, while bioresins must simultaneously satisfy requirements for photopolymerisation efficiency, cytocompatibility, and mechanical performance [120,121]. These constraints are particularly relevant to bone scaffolds containing high fractions of CaP, bioactive glass, or other solid mineral phases. Overall, VBP is currently better suited to the rapid fabrication of complex, cell-rich, and non-load-bearing bone-like tissues. For large load-bearing bone defects requiring high mineral content and substantial mechanical support, VBP will likely need to be integrated with material reinforcement strategies or hybrid manufacturing approaches.

5. Strategies by bioprinted materials promote osteogenesis through immunomodulation

During bone defect repair, the temporal transition, intensity, and spatial distribution of local immune responses directly influence inflammatory resolution, neovascularization, mesenchymal stem cell (MSC) recruitment, and subsequent bone matrix deposition, thereby serving as key determinants of osteogenic efficiency and osseointegration quality [122]. With the advancement of osteoimmunology, bone-repair materials are no longer regarded merely as passive structural supports. Instead, they are increasingly recognized as active regulators of the host immune microenvironment through multidimensional cues, including chemical composition, surface and structural characteristics, and bioactive signals. These cues can promote the transition of inflammatory responses toward regenerative repair and coordinate the coupling between vascularisation and osteogenesis [123]. Bioprinting provides an important technological foundation for constructing immunomodulatory bone-regenerative materials because of its capacity for high-precision spatial architectural control, multimaterial integration, and controlled delivery of bioactive factors [101]. Specifically, the chemical composition, structural and topographical features, mechanical properties, and loadable bioactive substances of bioprinted materials can all serve as important entry points for immune regulation. Natural or synthetic polymers may influence macrophage polarisation through surface functional groups, degradation products, and cell-adhesive microenvironments. Inorganic components or ion-releasing systems containing strontium (Sr), zinc (Zn), magnesium (Mg), calcium (Ca), and silicon (Si) can regulate inflammatory mediator expression and participate in pro-angiogenic and pro-osteogenic signaling [124]. Physical cues, such as pore size, pore interconnectivity, surface topography, and local curvature, can affect immune cell infiltration, cell morphology, and the distribution of immune phenotypes [13]. Meanwhile, printable delivery systems incorporating exosomes, cytokines, immunomodulatory small molecules, or nucleic acid carriers offer the potential to precisely intervene in inflammation initiation, resolution, and regenerative transition across both spatial and temporal dimensions [38]. Through these multidimensional regulatory strategies, bioprinted materials can actively shape the immunoregenerative microenvironment within bone defects, thereby optimising MSC homing, angiogenesis, osteogenic differentiation, and new bone maturation, and ultimately improving the quality and reproducibility of bone defect regeneration.

Therefore, this section discusses the design strategies of immunomodulatory bioprinted materials, focusing on three major dimensions: material-derived chemical cues, structural and topographical cues, and printable bioactive substances. It systematically summarizes the underlying mechanisms and recent advances by which these strategies regulate the local immune microenvironment and promote immune-vascular-osteogenic coordination, with the aim of providing a theoretical basis and practical reference for the future design of immunomodulatory bioinks and scaffold systems for bone regeneration.

5.1. Design principles for bioink systems

Immunomodulatory bone bioinks are not simple combinations of materials, cells, and bioactive components, but rather multiparametric systems in which printability, the mechanical microenvironment, degradation kinetics, cell loading, and manufacturing conditions interact with one another. Accordingly, their design should shift from optimising individual properties towards coordinated optimisation of multiple parameters [125,126]. Increasing polymer concentration, viscosity, or crosslinking density generally improves printing fidelity and structural stability, but may also increase extrusion induced shear stress and restrict cell spreading, migration, and matrix remodeling [125,127]. At the same time, the mechanical properties of the matrix can themselves be sensed by immune cells and subsequently influence cell adhesion, cytoskeletal organisation, and inflammatory state. Zhuang et al. found that GelMA hydrogels with different stiffnesses elicited distinct degrees of macrophage inflammatory activation, indicating that the mechanical microenvironment itself represents an important immunomodulatory variable [128]. Mechanical matching in bone bioprinting should therefore not be interpreted simply as increasing the modulus of cell laden hydrogels to approximate that of native bone. Instead, the matrix mechanics experienced at the cellular scale should be distinguished from the structural stability required at the construct scale. For load bearing bone defects, overall mechanical support can be provided by PCL, microfibrous frameworks, or inorganic reinforcing phases, while preserving sufficient compliance and remodellability within the cell laden hydrogel.

Degradation kinetics provide a further link among material mechanics, bioactive signal release, and host immune responses. Excessively rapid degradation may lead to premature loss of structural support and transient local accumulation of ions, drugs, or degradation products, thereby increasing inflammatory stimulation. By contrast, excessively slow degradation may restrict cellular and vascular infiltration and prolong the persistence of the foreign material interface. Wu et al. demonstrated that differences in the degradation behaviour of PCL/β TCP scaffolds markedly altered macrophage oxidative stress and inflammatory status, with subsequent effects on bone regeneration [129]. For bioinks containing bioactive ions, cytokines, exosomes, or nanocarriers, degradation rate should therefore be considered together with release kinetics. A more rational design should maintain sufficient structural stability and minimise burst release during the early post implantation period, followed by progressive degradation that creates space for cell infiltration, vascularisation, and new bone deposition, and ultimately allows gradual replacement of the material by regenerated tissue. Degradation behaviour, release kinetics, and immunomodulation should therefore be evaluated within a common temporal framework rather than treated as independent parameters.

Cell loading can in turn alter the rheological and printing behaviour of bioinks. Previous studies have shown that cell density can affect bioink viscosity, storage modulus, and gelation characteristics, indicating that cell concentration should not simply be introduced as an independent variable after optimisation of the acellular formulation [130,131]. Cell loading efficiency should likewise not be evaluated solely by short term post printing viability, but should also encompass actual encapsulation efficiency, spatial uniformity, long term survival, and retention of cellular function. For immunomodulatory bone bioinks, it is particularly important to determine whether printing induced shear stress, crosslinking conditions, and local mass transport alter the paracrine activity of MSCs, the inflammatory state of macrophages, or interactions among different cell populations. Thermal, radiation based, and chemical sterilisation methods may alter polymer chain structure, crosslinking behaviour, mechanical properties, degradation rate, and biological activity, and may therefore also affect subsequent immune responses. Rizwan et al. showed that different sterilisation treatments markedly altered the mechanical properties, degradation behaviour, and printability of GelMA [132]. Zhang et al. further demonstrated that sterilisation methods could modify macrophage related gene expression, indicating that sterilisation may represent an overlooked confounding factor in the evaluation of immunomodulatory materials [133]. For bioinks containing living cells, aggressive terminal sterilisation is generally not feasible after printing. Their manufacture therefore relies more heavily on sterile raw materials, aseptic preparation, sterile printing consumables, and controlled fabrication environments. Endotoxin control is particularly important because even low levels of contamination can activate macrophage inflammatory responses and confound interpretation of the intrinsic immunological effects of the material [134]. Overall, no single performance parameter of an immunomodulatory bone bioink can be maximised independently. Rational design should instead integrate printability, mechanical support, degradation and release behaviour, cellular function, sterilisation, and quality control according to the mechanical demands, inflammatory state, and vascularisation requirements of the target defect, thereby enabling material properties to adapt dynamically to the progression of bone repair rather than merely optimising any one parameter in isolation.

5.2. Material composition

Material composition defines the initial chemical interface formed between an implant and host tissue and therefore represents one of the most direct design dimensions through which bioprinted scaffolds can regulate the local immune microenvironment. Natural polymers, synthetic polymers, and inorganic components offer distinct advantages in terms of immunomodulatory capacity, structural support, engineering controllability, and clinical applicability, and no single material system is optimal for all types of bone defects [135]. Natural polymers generally exhibit favourable cytocompatibility and extracellular matrix mimetic properties, making them well suited for cell laden bioinks. They can also influence immune cell behaviour through adhesive ligands, degradation fragments, and matrix associated signalling. However, their mechanical performance, degradation stability, and batch to batch consistency are relatively limited, and additional reinforcement is often required when they are used for load bearing bone defects [29]. By contrast, the molecular composition, crosslinking degree, mechanical properties, and degradation behaviour of synthetic polymers can be more readily controlled, providing good processing reproducibility and structural support. Nevertheless, their intrinsic biological activity is usually limited, and the hydrophobic interfaces or local chemical changes associated with degradation of some synthetic materials may alter host inflammatory responses. They are therefore frequently combined with natural polymers, mineral phases, or other bioactive components [135]. Inorganic components and functional ions can simultaneously influence immune regulation, mineralisation, angiogenesis, and bone remodeling, giving them considerable potential for functional integration in bone tissue engineering. Their effects, however, are highly dependent on both concentration and exposure duration. In particular, functional ions such as Cu²⁺ and Zn²⁺ may induce cytotoxicity or oxidative stress at excessive concentrations, making release kinetics a critical determinant of both safety and therapeutic efficacy [136]. Accordingly, the central aim of material composition design should not be limited to identifying materials that promote M2 macrophage polarisation. Instead, it should establish an appropriate balance among the biological activity of natural polymers, the engineering controllability of synthetic polymers, and the functional properties of inorganic components according to the mechanical demands of the defect site, cell loading requirements, inflammatory status, infection risk, and degree of vascularisation.

5.2.1. Natural polymers

Natural polymers have attracted extensive attention in bone tissue engineering because of their favourable biocompatibility, biodegradability, and ability to recapitulate structural features of the native extracellular matrix (ECM) [137]. In addition to supporting cell adhesion and differentiation, these materials can modulate immune responses and thereby influence macrophage functional states and subsequent osteogenesis. Immune regulation is particularly important during bone regeneration because excessive or persistent inflammation can impair healing, whereas appropriately controlled immune responses can support tissue regeneration and osseointegration [138,139]. Mineralised collagen provides a representative example. Studies using non-printed materials have shown that its microarchitecture and mineral composition can influence macrophage recruitment and the development of repair-associated phenotypes, with subsequent effects on MSC recruitment and osteogenesis [140]. Although these studies did not involve bioprinting, they reveal fundamental mechanisms through which natural matrix components participate in osteoimmune regulation and therefore provide a mechanistic basis for the compositional design of natural polymer-based bioinks.

When these materials are further incorporated into three-dimensional printing systems, material chemistry and three-dimensional architecture can act together to shape the local immune microenvironment. Patel et al. developed printed composite scaffolds comprising chitosan, silk fibroin, and cellulose nanoparticles. Incorporation of cellulose nanoparticles and silk fibroin improved the rheological and printability characteristics of the chitosan-based system, while promoting a shift in macrophage phenotype from an M1-associated towards an M2-associated state and subsequently enhancing bone regeneration [141]. In true three-dimensional bioprinting systems, an additional advantage of natural polymers is their ability to simultaneously encapsulate living cells and functional components. Sun et al. formulated a composite bioink consisting of GelMA, SilMA, and gelatin-dopamine, further incorporating graphene oxide and BMSCs to fabricate an artificial periosteum with tissue adhesive properties (Fig. 4A) [142]. The system maintained favourable cell viability and printing stability, while graphene oxide further enhanced BMSC osteogenic differentiation and in vivo bone formation (Fig. 4B). More direct evidence of immunomodulation was provided by the GelMA/Sr-CSH composite bioink developed by Yu et al. [10]. In this system, GelMA served as the cell-laden matrix, while Sr-CSH nanowires were incorporated into the bioink together with BMSCs. Sr-CSH not only improved printing fidelity and mechanical performance, but also promoted repair-associated macrophage polarisation and inflammatory resolution through the release of bioactive ions, thereby enhancing BMSC osteogenic differentiation and producing pronounced regeneration in a rat critical-sized calvarial defect model. Collectively, these findings suggest that the role of natural polymers in immunomodulatory bioprinting is better understood as that of an engineerable matrix platform. The polymers themselves provide an ECM-like cellular microenvironment, whereas more directional immunomodulatory effects often require synergistic incorporation of inorganic nanocomponents, bioactive ions, or other functional agents.

Fig. 4.

A Immunofluorescence staining showing increased fluorescence intensity of the osteogenic marker RUNX2 in BMSCs within the 3D-bioprinted artificial periosteum. B, Histological evaluation of the calvarial defect region at 8 weeks after surgery, including low- and high-magnification H&E staining, Masson's trichrome staining, and immunofluorescence staining of RUNX2 and OCN [142]. Copyright 2023, IOP Publishing Ltd. C, Fabrication of immunomodulatory scaffolds using a composite bioink consisting of gelatin methacryloyl (GelMA), alginate, and four-arm polyethylene glycol acrylate (PEG acrylate), combined with coaxial bioprinting and a sequential crosslinking strategy. D, Histological evaluation of bone repair in different groups at 8 weeks after implantation using H&E and Masson's trichrome staining [17]. Copyright 2022, Elsevier.

Accordingly, the design of natural polymer-based bioinks should extend beyond improving cytocompatibility or pursuing a single M2-associated macrophage phenotype. Greater attention should be given to the relationships among matrix origin, degradation behaviour, incorporation of functional components, and the host immune response after implantation. Only by distinguishing the intrinsic immunocompatibility of the material from its actively engineered immunomodulatory functions can the actual contribution of natural polymers to immunomodulatory osteogenic bioprinting be assessed more accurately.

5.2.2. Synthetic polymers

Unlike natural polymers, which influence cell behaviour largely through intrinsic adhesion sites and matrix derived biological signals, synthetic polymers are distinguished by the ability to independently tune their molecular architecture, functional group density, crosslinking degree, degradation profile, and mechanical properties across a relatively broad range. This enables the construction of cellular microenvironments with more precisely defined compositions and greater reproducibility [143]. Such engineering controllability also means that the immune effects of synthetic polymers rarely arise from an inherent and fixed anti inflammatory or pro reparative activity of the polymer itself. Instead, they are jointly determined by multiple parameters, including protein adsorption, cell adhesion, network stiffness, ligand density, and degradation behaviour [144]. Polyethylene glycol (PEG) hydrogels provide a representative example. Although PEG generally exhibits low nonspecific protein adsorption and limited intrinsic cell adhesion, this does not imply an absence of host immune responses after implantation. Lynn et al. found that unmodified PEG hydrogels elicited a pronounced macrophage inflammatory response, whereas incorporation of RGD reduced proinflammatory signals such as TNF-α and IL-1β, indicating that modification of the cell adhesion interface alone can substantially alter the host response [145]. Blakney et al. further demonstrated that, even when RGD density was maintained at a constant level, PEG hydrogel stiffness influenced macrophage morphology and the degree of inflammatory activation, with softer networks inducing lower inflammatory cytokine expression and a reduced long term foreign body response [146]. Accordingly, the immunomodulatory value of synthetic polymers lies primarily in their capacity to convert material chemistry and physical properties into tunable design variables, rather than in categorising any particular polymer as intrinsically proinflammatory or anti inflammatory.

Early cell bioprinting studies already reflected this design concept. Gao et al. used inkjet bioprinting to deposit an acrylatable peptide together with a PEG hydrogel while simultaneously incorporating human bone marrow derived MSCs. By introducing adhesive peptide sequences into the PEG network, the system improved interactions between cells and the synthetic matrix while maintaining printing fidelity and subsequently supported both osteogenic and chondrogenic differentiation [147]. Ma et al. subsequently combined PEGDMA with GelMA to develop a photocrosslinkable bioink containing periodontal ligament stem cells and systematically varied the GelMA to PEGDMA composition ratio. Different network compositions markedly altered cellular behaviour, with the GelMA/PEGDMA formulation at a ratio of 4:1 exhibiting favourable osteogenic differentiation and producing greater alveolar bone formation after implantation [148]. These studies demonstrate that the initial value of synthetic polymers in bioprinting lies in their ability to quantitatively regulate the three dimensional network surrounding cells. However, these early studies primarily focused on cytocompatibility and osteogenesis and did not directly evaluate immune processes. PEG itself should therefore not be regarded as an immunomodulatory factor on the basis of these findings alone.

With the incorporation of additional functional components, PEG based bioinks have evolved from serving primarily as controllable three dimensional matrices towards systems that simultaneously support cell delivery and microenvironmental regulation. Zhai et al. developed a dual bioink system composed of PEGDA, Laponite, and hyaluronic acid and used dual channel three dimensional bioprinting to integrate rat osteoblasts into the constructs [149]. PEGDA contributed to the formation of a stable crosslinked network, whereas Laponite improved material performance and provided sustained release of Mg and Si related bioactive ions. Post printing cell viability exceeded 95% in the short term, and the constructs subsequently exhibited favourable osteogenic performance both in culture and in vivo. A representative study directly linking a synthetic polymer network with osteoimmune regulation was reported in a multicellular bioprinting system. Sun et al. developed a composite bioink consisting of GelMA, gelatin, and four arm PEG acrylate, in which BMSCs and RAW264.7 macrophages were coencapsulated together with mesoporous silica nanoparticles loaded with BMP-4 [150]. The four arm PEG acrylate enhanced the mechanical stability of the hydrogel network, allowing different cell populations and bioactive components to be retained within the printed construct. Sustained BMP-4 release promoted the development of a CD206 associated reparative macrophage phenotype while reducing proinflammatory mediator levels. In a critical sized calvarial defect model in diabetic rats, this multicellular bioprinted system ameliorated the dysregulated inflammatory microenvironment and accelerated bone repair. Building on this concept, synthetic polymer networks have also been used to regulate the temporal sequence of immune signalling. Ma et al. formulated a composite bioink containing GelMA, alginate, and four arm PEG acrylate and used coaxial bioprinting with sequential crosslinking to fabricate microfibrous structures, while Laponite was employed as a carrier for IFN-γ (Fig. 4C) [17]. Initial release of IFN-γ promoted an early M1 associated immune response, after which the gradual release of Mg and Si related ions from Laponite drove macrophages towards a more reparative state. In parallel with this temporal immune transition, the printed constructs promoted endothelial progenitor cell migration, vascular formation, and subsequent vessel maturation, ultimately enhancing vascularized bone regeneration in a calvarial defect model (Fig. 4D).

5.2.3. Inorganic ions

Inorganic ions are essential components of the mineralised microenvironment of bone. Their roles in bone regeneration extend beyond supplying the chemical elements required for mineralisation, as they can also participate in inflammatory resolution, angiogenesis, and bone remodeling by influencing immune cells, mesenchymal stem cells, endothelial cells, and osteoclasts [151]. Ca²⁺, PO₄³⁻, and Si related ionic species primarily contribute to mineral deposition and maintenance of an osteogenic microenvironment, while their local concentrations, dissolution forms, and material surface reactivity can also alter the responses of macrophages and other host cells [152,153]. Functional ions such as Mg²⁺, Sr²⁺, and Zn²⁺ further participate in the regulation of inflammatory signalling and bone metabolism [154]. Qiao et al. demonstrated that the effects of Mg²⁺ on bone repair are strongly stage dependent. During the early phase, Mg²⁺ can establish an osteogenesis supporting immune environment through macrophage TRPM7 related signalling, whereas prolonged exposure may enhance NF κB activity, promote the formation of osteoclast like cells, and delay bone maturation [155]. Zn and Sr codoped hydroxyapatite has likewise shown immunomodulatory and osteogenic effects that vary over the course of repair [124]. The biological effects of inorganic ions therefore cannot be simply summarised as promoting M2 polarisation or osteogenesis. Their ultimate effects depend on ion identity, local concentration, exposure duration, and the stage of bone repair. For bone regenerative materials, controlling the pattern of ion exposure is therefore as important as selecting the ion itself.

This consideration is particularly important in three dimensional bioprinting. Conventional inorganic scaffolds can often provide relatively high inorganic contents through bulk material dissolution, whereas cell laden bioinks must first preserve cellular viability during and after printing. As a result, the amount, particle morphology, and ion release rate of incorporated inorganic phases are subject to considerably tighter constraints. Once dispersed within hydrogels, inorganic particles can also alter bioink viscosity, shear thinning behaviour, crosslinking, and shape retention, giving them both processing and biological functions. Dubey et al. incorporated amorphous magnesium phosphate into an extracellular matrix mimetic hydrogel to develop an ECM/AMP bioink containing dental pulp stem cells. The addition of AMP maintained approximately 90% cell viability after printing and enhanced mineralisation and osteogenic gene expression without the addition of exogenous osteogenic growth factors [156]. Subsequently, Mg²⁺ doped nHA was incorporated into a gelatin/GelMA bioink for printing hBMSCs. nHA with different crystallinities, morphologies, and ionic substitution characteristics exerted distinct effects on extrudability, structural retention, cellular metabolism, and osteogenic differentiation, indicating that the effects of inorganic phases cannot be predicted solely from elemental composition. Crystal phase and particle characteristics are also important determinants of cellular responses after bioprinting [157]. However, these studies primarily evaluated printability and osteogenesis and did not directly establish macrophage mediated immune mechanisms. They are therefore more appropriately regarded as foundational evidence for ion functionalised bioinks rather than as direct demonstrations of immunomodulatory bioprinting.

Building on these findings, recent studies have begun to directly integrate ion release with host immune regulation in cell laden bioinks. Liu et al. developed a bdECM MA/Si CaP bioink composed of methacrylated bone derived decellularised extracellular matrix and silicon substituted calcium phosphate and encapsulated BMSCs for DLP three dimensional bioprinting (Fig. 5A) [11]. The system maintained high cell viability, achieved mechanical properties at the MPa scale, and simultaneously exhibited osteogenic, angiogenic, and immunomodulatory effects. Importantly, the in vivo immune response did not simply involve sustained promotion of an anti inflammatory phenotype, but instead displayed a temporal transition from an early proinflammatory response towards a later anti inflammatory state, with suppression of p38 MAPK signalling contributing to subsequent immune regulation (Fig. 5B and C). Bioprinting studies involving Sr based inorganic phases further illustrate the progression from simple enhancement of osteogenesis towards coordinated osteoimmune regulation. In an earlier study, Alcala Orozco et al. incorporated Sr based nanoparticles into low concentration GelMA to develop a Sr GelMA bioink containing hMSCs. Sr incorporation increased bioink viscosity and shape retention while maintaining post printing cell viability above 95%, and also enhanced ALP, OCN, and COL I expression together with matrix mineralisation [158]. This study primarily established the effects of Sr functionalisation on printability and osteogenesis, but did not directly assess immune regulation. Yu et al. subsequently advanced this strategy by incorporating 5% Sr substituted calcium silicate hydrate nanowires into 10% GelMA and encapsulating BMSCs to generate a GelMA/Sr CSH bioink [10]. Sr CSH improved both printing fidelity and mechanical performance and, through sustained release of Sr, Ca, and Si related bioactive ions, also modulated macrophage responses, increasing CD206 associated reparative phenotypes and improving the local inflammatory environment. This immune modulation was accompanied by enhanced osteogenic differentiation of BMSCs and ultimately by substantial new bone formation in a rat critical sized calvarial defect model. Compared with approaches that use Sr merely as an osteogenic additive, this study established a more complete link among inorganic phase dissolution, macrophage regulation, BMSC osteogenesis, and in vivo bone regeneration, making it a representative example of ion mediated immunomodulatory bone bioprinting. The same cell laden GelMA/Sr CSH system was subsequently applied to biphasic multicellular osteochondral bioprinting, in which Sr CSH was spatially confined to the BMSC laden subchondral bone region and promoted osteogenic differentiation through sustained release of bioactive ions [159]. Zn based systems represent another direction in which ion release is integrated with drug delivery within bioinks. Yu et al. incorporated luteolin loaded ZIF 8 nanoparticles into GelMA and further encapsulated BMSCs for three dimensional bioprinting [12]. During ZIF 8 degradation, Zn²⁺ was continuously released together with luteolin, which possesses anti inflammatory and antioxidant properties. Compared with GelMA alone, this system reduced macrophage iNOS expression, increased CD206 expression, and suppressed proinflammatory signalling, thereby establishing a microenvironment more favourable for BMSC osteogenic differentiation. This was subsequently reflected by increased RUNX2, ALP, COL1A1, and OCN expression and enhanced matrix mineralisation. In vivo calvarial defect experiments further showed that the system reduced early inflammation and promoted new bone formation.

Fig. 5.

A, Schematic illustration of the design and fabrication of 3D-bioprinted Si-CaP-loaded tissue-engineered bone, exhibiting favourable mechanical properties, bioactivity, and immunomodulatory potential. B, Immunofluorescence staining of iNOS at the implantation site of Si-CaP-loaded bioprinted scaffolds at weeks 1, 2, and 3. Green and blue indicate iNOS and cell nuclei, respectively. The lower panels show magnified views of the boxed regions in the corresponding upper images. C, Immunofluorescence staining of CD163 at the implantation site of Si-CaP-loaded bioprinted scaffolds at weeks 1, 2, and 3. Red and blue indicate CD163 and cell nuclei, respectively. The lower panels show magnified views of the boxed regions in the corresponding upper images [11]. Copyright 2024, Wiley. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Overall, the role of inorganic ions in three dimensional bone bioprinting is evolving from enhancement of mineralisation and osteogenesis towards active regulation of the immune microenvironment, although the maturity of evidence varies considerably among different ions. Mg based and Ca/P based bioinks are supported by substantial evidence regarding cytocompatibility, printability, and osteogenic activity, but direct osteoimmune studies remain relatively limited. Si CaP and Sr CSH systems have begun to establish links among temporal immune responses, macrophage states, and subsequent osteogenesis, whereas Zn based MOF systems further integrate ion release with delivery of other immunologically active components. Future research should therefore move beyond simply determining whether a bioink contains a particular functional ion and instead focus on controlling release dose, duration, and spatial distribution. Experimental designs should also distinguish the respective contributions of the inorganic particles themselves, released ions, and other coencapsulated bioactive components. Multimaterial bioprinting can assign distinct inorganic compositions to different regions within a single construct. When combined with controlled degradation and stage specific release, this approach may enable ionic signals to be more precisely matched to different phases of repair, including inflammatory initiation, inflammatory resolution, vascularisation, and mineralisation.

5.3. Surface topography

Unlike strategies that directly modulate immune signalling through material chemistry or exogenous bioactive molecules, scaffold architecture and surface topography primarily regulate immune responses through cell–material interactions [160], mass transport, and mechanotransduction. A major advantage of this approach is that immunomodulatory functions can be encoded directly into scaffold geometry without the need for continuous delivery of drugs or cytokines, thereby offering favourable spatial stability and manufacturing reproducibility [135,161]. Pore size, porosity, and interconnectivity mainly operate at the tissue scale by regulating immune cell infiltration, oxygen and nutrient diffusion, vascular ingrowth, and tissue occupancy, whereas local curvature and micro/nanoscale topography act predominantly at the cellular scale by influencing focal adhesion formation, cytoskeletal reorganisation, and mechanosensitive signalling, thereby altering the functional states of macrophages and other host cells [161,162]. However, structural regulation does not involve any single universally optimal parameter. Although increasing pore size or porosity generally facilitates cellular and vascular infiltration, it may simultaneously reduce specific surface area and scaffold mechanical strength. Moreover, pore size, porosity, curvature, and permeability are highly interdependent, making it difficult to fully isolate the biological effects of any individual parameter [161]. Differences in material type, printing resolution, and observation period may also produce distinct or even conflicting conclusions regarding optimal pore dimensions or topographies. Therefore, a fixed relationship should not be assumed between any single geometric parameter and a specific immune phenotype [162,163]. Overall, structural regulation is particularly suitable for bone scaffolds intended for long term implantation, reduced reliance on exogenous bioactive factors, or cell free regenerative strategies. However, because scaffold geometry alone has limited capacity to adapt dynamically to stage specific immune changes during bone healing, it is better regarded as a foundational physical regulatory framework that should be integrated with material chemistry and controlled delivery of biological signals.

5.3.1. Structural parameters

5.3.1.1. Pore size and interconnectivity

Pore size and pore interconnectivity are key structural parameters governing mass transport, cell migration, and tissue ingrowth within three-dimensional constructs. Open and interconnected pores facilitate the diffusion of oxygen and nutrients, removal of metabolic waste, infiltration of host cells, and establishment of vascular networks [164,165]. From an immunomodulatory perspective, pore architecture can also influence macrophage function by altering the space available for immune cell migration, cell spreading, and local mass transport conditions. Previous studies have shown that spatial confinement can affect macrophage cytoskeletal organisation and associated phenotypic signalling, whereas open porous networks facilitate immune cell infiltration and tissue remodeling [143,166]. Pore architecture can therefore shape the local immune microenvironment both directly, by altering cell–material contact, and indirectly, through its effects on oxygen supply, cell migration, and vascularisation.

These effects are more complex in cell-laden three-dimensional bioprinting. Conventional acellular porous scaffolds primarily rely on host cells infiltrating from the periphery after implantation, whereas bioprinted constructs may already contain MSCs, immune cells, or endothelial cells at the time of fabrication. Pore architecture therefore also determines the survival, migration, and intercellular communication of the printed cells. This is particularly relevant to hydrogel bioinks. Although a higher degree of crosslinking generally improves shape retention, it can generate dense polymer networks that restrict cell spreading and molecular diffusion, whereas reducing crosslinking may compromise printing fidelity and structural stability [167]. Pore design in bioprinting must therefore balance fabrication stability, cell-accessible remodeling space, and effective mass transport. To overcome restricted cell migration within dense hydrogels, Tao et al. encapsulated BMSCs in a GelMA/dextran aqueous emulsion for DLP bioprinting and subsequently removed the dextran phase to generate open internal pores [167]. Compared with dense GelMA, the porous construct promoted BMSC spreading, migration, and proliferation, enhanced YAP nuclear translocation and osteogenic differentiation, and improved calvarial defect repair. This study demonstrates that pore design in bioprinting extends beyond the macroscopic lattice generated by the printing path and can also include secondary pores within cell-laden printed filaments. Although immune cells were not directly evaluated, such open internal architecture provides the spatial basis required for host macrophages and other reparative cells to infiltrate the printed region after implantation. In larger constructs, continuity between pores becomes increasingly important for long-range mass transport and vascularisation. Shen et al. used dual-nozzle bioprinting to generate continuous tubular channels of approximately 500 μm within BMSC-laden GelMA and positioned endothelial cells along the channel surfaces, thereby promoting vascular network formation and subsequent angiogenic–osteogenic coupling (Fig. 6A and B) [16]. Bioprinting can also exploit printing trajectories to regulate the spatial relationships among different cell populations. Yu et al. separately encapsulated bone marrow-derived macrophages and BMSCs in two bioinks and used dual-channel printing to organise the two cell types within the same construct. Paracrine interactions between these populations promoted an early repair-associated macrophage phenotype and subsequently enhanced BMSC osteogenesis and calvarial defect repair (Fig. 6C and D) [168]. These findings indicate that, in bioprinted systems, channels serve not only as pathways for mass transport but can also contribute to osteoimmune regulation by organising the spatial proximity of distinct cell populations.

Fig. 6.

A, Schematic illustration of 3D bioprinting of BMSC-laden GelMA constructs containing continuous channels of approximately 500 μm to promote immunomodulation, vascularisation, and osteogenesis. B, Micro-CT evaluation of bone defect repair, showing three-dimensional reconstructed images of calvarial defects at 1, 4, 8, and 12 weeks after surgery [16]. Copyright 2022, Elsevier. C, Schematic illustration of dual-channel bioprinting in which bone marrow-derived macrophages (BMMs) and BMSCs were separately encapsulated within the same construct to promote bone repair through early immunomodulation and subsequent osteogenic induction. D, H&E and Masson's trichrome staining of the different experimental groups at 1 and 2 months after surgery [168]. Copyright 2022, ACS.

Accordingly, the roles of pore size and interconnectivity in immunomodulatory three-dimensional bioprinting should be understood within a multiscale spatial framework. Macroscopic printed channels primarily regulate tissue and vascular infiltration, pores within the hydrogel matrix influence the spreading and matrix remodeling of encapsulated cells, and printing trajectories can further control spatial interactions among immune cells, MSCs, and endothelial cells [16,165]. This also means that optimal pore sizes identified in conventional acellular PCL, PEEK, or bioceramic scaffolds cannot be directly extrapolated to cell-laden bioprinted systems, because bioink crosslinking, degradation, cell density, and mechanical properties all alter the actual cellular microenvironment associated with a given pore architecture.

5.3.1.2. Porosity

Porosity refers to the proportion of void volume within a construct and is therefore distinct from pore size, which describes the dimensions of individual pores. An appropriate porosity can increase the space available for cells and tissue, improve mass exchange, and facilitate host tissue infiltration, whereas excessively high porosity may compromise scaffold stiffness and structural stability. Its design therefore requires a balance between biological requirements and mechanical performance [169,170]. For cell-laden bioprinted constructs, it is also important to distinguish between macropores formed between printed filaments and micropores within the bioink itself. The former primarily govern overall tissue ingrowth and vascular invasion, whereas the latter directly affect the spreading, migration, and matrix remodeling of encapsulated cells.

Such internal porosity is particularly important for hydrogel bioinks. Although highly crosslinked hydrogels are advantageous for maintaining printing fidelity, their dense internal networks can restrict cell migration and cell–cell contact. Zauchner et al. introduced microporosity into PEG hydrogels and showed that open, cell-scale pores promoted rapid spreading of MSCs and osteoblasts, formation of three-dimensional cellular networks, and enhanced matrix deposition and osteogenic differentiation [171]. Koo et al. used a highly porous collagen/HA bioink to extrusion print hASC-laden constructs containing hierarchical macro- and microporosity within the cell-laden filaments [172]. Compared with denser constructs, this system improved cell–cell interactions and mass transport and promoted cell proliferation together with increased expression of osteogenic markers including COL1, BMP2, and OCN. Similarly, porous collagen/β-TCP bioinks have been used for bioprinting hASCs alone or together with HUVECs, supporting osteogenesis and vascular network formation while maintaining cell loading [173,174]. These studies suggest that increasing effective porosity within bioinks can alleviate the constraints imposed by dense hydrogel networks on cellular activity and create a more favourable spatial environment for vascularisation and tissue remodeling.

Studies of conventional porous implants have already demonstrated that pore architecture can influence macrophage infiltration and the foreign body response. In cell-laden bone bioprinting, however, most studies still focus primarily on cell viability, osteogenesis, and vascularisation. Appropriate porosity should therefore be regarded as providing a structural basis for immunomodulation by facilitating cellular remodeling and host tissue infiltration, rather than as a parameter that necessarily induces a specific macrophage phenotype by itself.

5.3.2. Surface morphology/topology

Pore size, porosity, and interconnectivity primarily regulate spatial transport and cellular infiltration at the tissue scale, whereas surface morphology and topography act directly at the material interface encountered by cells, with their biological effects occurring predominantly at the cellular scale. Curvature, roughness, grooves, and micro- and nanoscale patterns can alter the local contact state between the cell membrane and material surface and subsequently regulate focal adhesion distribution, cytoskeletal tension, nuclear deformation, and mechanosensitive signalling [175]. These parameters therefore do not primarily determine whether cells can enter the scaffold, but rather how cells perceive and respond to the local interface after infiltration.

Curvature is one of the more readily quantifiable topographical parameters and can be characterised using metrics such as mean curvature and Gaussian curvature [175]. Pieuchot et al. demonstrated that adherent cells can sense continuously varying curvature fields at the cellular scale and preferentially migrate towards specific concave regions through mechanical coupling among focal adhesions, the actin cytoskeleton, and the nucleus [176]. Callens et al. further showed that the spatial distribution, stress fibre orientation, and extracellular matrix organisation of preosteoblasts within complex curvature fields were strongly curvature dependent [177]. These findings indicate that curvature is not merely a geometric descriptor, but can be converted into a biologically active mechanical signal by altering cellular force transmission and spatial organisation. Similar interfacial mechanical effects also apply to immune cells. Hu et al. fabricated silk fibroin membranes with different surface roughness while maintaining other material properties as consistently as possible and found that rougher surfaces induced local deformation of the macrophage membrane, promoted integrin αv internalisation, and suppressed integrin/NF-κB-related signalling. These changes attenuated the early inflammatory response and shifted macrophages towards a repair-associated state [178]. This finding demonstrates that the immunomodulatory effects of surface morphology do not necessarily depend on the release of soluble factors, but can instead be transmitted directly from nanoscale and microscale interfacial deformation to cell adhesion and inflammatory signalling.

Nanoscale surface patterning further illustrates this scale-dependent regulation [14]. Zhu et al. generated honeycomb TiO₂ structures of different feature sizes on titanium surfaces and observed marked changes in macrophage responses as the characteristic dimension decreased. Among these structures, the 90 nm honeycomb pattern most prominently increased CD206, IL-4, and IL-10 expression together with BMP-2 secretion. Transcriptomic analysis further suggested that this nanotopography promoted macrophage pseudopod formation and activated cytoskeleton-related regulators including RhoA, Rac1, and CDC42. The RhoA/ROCK pathway contributed to macrophage state regulation, ultimately enhancing MSC osteogenesis and implant osseointegration [102]. This study indicates that nanoscale patterns can convert geometric information into immune signals by controlling adhesion structures and cytoskeletal organisation, rather than acting merely through an increase in surface area. Micro- and nanoscale features generated by surface coatings can participate in similar processes. Shi et al. deposited a CaP coating onto highly ordered MEW-PCL scaffolds and found that the coating not only increased surface roughness and hydrophilicity, but that surface morphology and Ca²⁺ release were respectively associated with PI3K/AKT- and cAMP-PKA-related signalling. Together, these effects promoted a repair-associated macrophage state and subsequently enhanced BMSC osteogenesis and in vivo bone formation [15]. This finding also highlights that surface morphology, surface chemistry, and ion release are rarely completely independent variables in practical scaffold systems. Accordingly, observed immune effects should not be attributed entirely to roughness or to any single nanoscale feature. Curvature and surface topography can also be directly encoded into cell-laden constructs through high-resolution bioprinting. Yang et al. used multimaterial DLP bioprinting to fabricate hydrogel structures containing helical grooves and TPMS surface topographies, demonstrating that precisely defined surface microstructures can regulate cell alignment, spreading, and migration [179]. Han et al. further fabricated high-density cell-laden GelMA Gyroid hydrogels in which continuous Gaussian curvature promoted endothelial cell self-organisation and formation of dense microvascular networks, showing that complex topographies can further influence cytoskeletal organisation and vascular patterning within cell-laden environments [180]. For osteoimmunomodulation, these fabrication capabilities suggest that curvature and micro- and nanoscale surface features could ultimately be designed together with the spatial organisation of immune, osteogenic, and endothelial cells, rather than being restricted to the surfaces of acellular implants. In the context of bone repair, Lu et al. provided more direct quantitative evidence that curvature can regulate osteoimmune responses. Using three-dimensional printed β-TCP scaffolds with different Gaussian curvatures, they found that negative Gaussian curvature suppressed macrophage Ras-MAPK/HIF-1α signalling and increased BMP-2 and VEGF secretion, thereby promoting vascularized bone formation [14]. Distinct curvature ranges produced different macrophage and osteogenic responses, indicating that curvature also operates within a quantitative response window and that concave surfaces or negative curvature should not be regarded as universally favourable structural features.

5.4. Immunoregulatory bioactive components by functional mode

Material composition and structural cues primarily shape immune responses indirectly through interactions at the material–host interface, whereas immunologically active components such as stem cells, cytokines, exosomes, and dECM can provide stronger biological signals and enable more active regulation of immune, vascular, and osteogenic coupling. However, this increased biological potency is accompanied by greater functional complexity and translational challenges [181]. Stem cells can sense the local inflammatory environment and dynamically regulate macrophages, lymphocytes, endothelial cells, and osteogenic cells through paracrine signalling, while also serving as a potential source of osteogenic cells. They are therefore particularly suitable for complex defects requiring coordinated regulation of multiple cell populations. Their therapeutic performance, however, is strongly influenced by donor variability, expansion status, printing induced shear stress, post implantation survival, and manufacturing and storage conditions [104,181]. Cytokines act through relatively well defined molecular targets, and their dose and release sequence can be more readily engineered, making them suitable for stage specific or programmed immunomodulatory strategies. Nevertheless, their short half lives and narrow therapeutic windows remain important limitations, while high doses or prolonged nonphysiological exposure may produce off target effects. A single cytokine is therefore unlikely to fully reproduce the complex signalling networks involved in physiological bone repair [181]. Exosomes combine, to some extent, the multitarget paracrine effects of cells with the engineering advantages of cell free therapy and can be incorporated into bioinks to achieve local retention and sustained release. However, their cargo composition varies substantially with donor source, cellular state, and preparation method, and major challenges remain in isolation and purification, dose standardisation, potency assessment, and scalable manufacturing [182]. dECM provides a multicomponent regulatory environment that more closely resembles native tissue by retaining tissue specific matrix proteins, adhesion sites, and matrix bound signals, making it particularly attractive for bioink designs that emphasise microenvironmental biomimicry. Its immunomodulatory effects are relatively indirect, however, and are strongly influenced by tissue source, the extent of decellularisation, sterilisation method, and batch variability. Its relatively weak rheological and mechanical properties also limit its use as a stand alone printable material [183]. These immunologically active components should therefore not be ranked simply according to their osteogenic capacity. Stem cells are better suited to dynamic multicellular regulation, cytokines to defined and programmable signalling interventions, exosomes to cell free multitarget regulation, and dECM to reconstruction of tissue specific microenvironments. Their selection should ultimately depend on the required regulatory precision and duration, defect complexity, and the acceptable manufacturing and regulatory burden.

5.4.1. Live cell-mediated immunoregulation: stem cells

In immunomodulatory bioprinting systems for bone regeneration, bone marrow mesenchymal stem cells (BMSCs) and adipose derived mesenchymal stem cells (ADSCs) function not only as osteogenic cell sources but also as active components that participate in local immune regulation [184,185]. MSCs can sense the inflammatory microenvironment and modulate immune cell functions through paracrine signalling and direct cell contact. Németh et al. found that inflammatory stimulation induced bone marrow stromal cells to promote macrophage IL-10 production through prostaglandin E2 (PGE2) [186]. Liu et al. further demonstrated that MSC derived TGF-β regulated the transition of macrophages towards an M2 like phenotype through the Akt/FoxO1 pathway [187], while CD54 dependent direct contact between MSCs and proinflammatory macrophages enhanced the immunosuppressive activity of MSCs [188]. These findings indicate that interactions between MSCs and macrophages represent a bidirectional regulatory process shaped by local inflammatory signals, rather than a unidirectional process in which MSCs simply induce M2 polarisation.

More importantly, changes in macrophage phenotype cannot be equated directly with enhanced osteogenesis. The regulation of MSC osteogenic fate by immune cells depends on specific paracrine mediators and their downstream signalling networks. Sun et al. developed a multicellular bioprinted scaffold containing BMSCs, RAW264.7 macrophages, and BMP-4 releasing nanoparticles. BMP-4 promoted a reparative macrophage phenotype and increased macrophage BMP-2 expression, while simultaneously enhancing osteogenic markers in BMSCs, including RUNX2, OPN, and ALP [150]. These findings suggest that BMP signalling may represent an important mechanistic link between macrophages and MSCs. Further mechanistic studies have shown that small extracellular vesicles derived from material modulated macrophages can activate the BMP2/Smad5 pathway in MSCs and enhance osteogenic differentiation, whereas inhibition of vesicle release markedly attenuates this effect [189]. In addition to BMP/Smad signalling, β-catenin related pathways also participate in osteoimmune coupling. Zhao et al. found that the macrophage MSR1 mediated PI3K/AKT/GSK3β/β-catenin pathway contributed both to maintenance of an M2 like metabolic state in macrophages and to their capacity to promote osteogenic differentiation of BMSCs [190]. An M2 like phenotype should therefore be regarded as one indicator of immune status rather than as a direct and sufficient condition for osteogenesis.

A major advantage of bioprinting is its ability to further regulate the spatial organisation of these cellular interactions. Hydrogel bioinks can serve as three dimensional delivery matrices for MSCs, enabling cell encapsulation during printing while preserving cell viability (Fig. 7A). The encapsulated MSCs can simultaneously modulate the local immune response through paracrine signalling, suppress excessive post implantation inflammation, and promote a transition of macrophages from an M1 associated towards an M2 associated phenotype. This can establish an immune microenvironment that supports bone repair, further enhance MSC osteogenic differentiation, and promote bone defect regeneration (Fig. 7B and C) [12]. Yu et al. used dual channel bioprinting to encapsulate bone marrow derived macrophages and BMSCs separately within different bioinks. BMSCs promoted an early transition of macrophages towards a reparative phenotype, while macrophages reciprocally enhanced osteogenic differentiation of BMSCs. In a rat calvarial defect model, the dual cell constructs achieved superior bone regeneration compared with single cell constructs [168].

Fig. 7.

A, Schematic illustration of a bioprinted scaffold loaded with bone marrow mesenchymal stem cells (BMSCs) for bone regeneration. B, The stem cell-laden bioprinted scaffold suppresses M1 macrophage polarisation and promotes M2 macrophage polarisation, thereby reducing inflammatory activation. Inducible nitric oxide synthase (iNOS) represents the M1 macrophage phenotype, and CD206 represents the M2 macrophage phenotype. C, Immunofluorescence staining of iNOS and CD206 in the stem cell-laden bioprinted scaffold [12]. Copyright 2025, Elsevier.

At the same time, the M1/M2 dichotomy itself represents only a simplified framework for describing macrophage states. Macrophages involved in bone repair exhibit substantial continuity, plasticity, and subset heterogeneity. Different M2 like states may display distinct functional characteristics, including those conventionally described as M2a to M2d, and their roles vary across different stages of repair [87]. During the early phase of injury, proinflammatory macrophages contribute to necrotic tissue clearance, immune cell recruitment, and initiation of regenerative signalling, while mediators such as PGE2 and OSM can also support early osteogenic commitment. As inflammation resolves, reparative macrophages become increasingly involved in angiogenesis, matrix deposition, and mineralisation. Immunomodulatory bioprinting should therefore not aim simply to sustain high CD206 expression or continuously increase the M2/M1 ratio. Instead, cell proportions, spatial organisation, bioink composition, and the temporal release of bioactive factors should be coordinated to promote dynamic immune transitions and activation of specific osteogenic signals that are appropriate to each stage of bone repair [[191], [192], [193]]. Such spatiotemporal regulation may more closely reproduce physiological bone healing than the pursuit of a single static M2 macrophage state. Nevertheless, differences in MSC source, replicative senescence during in vitro expansion, printing induced shear damage, and insufficient long term cell survival remain important challenges that must be addressed before these strategies can be translated further [192,193].

5.4.2. Soluble signal-mediated immunoregulation: cytokines

Stem cells regulate local immunity through dynamic paracrine networks and intercellular interactions, whereas cytokines act through relatively well defined receptors and downstream signalling pathways to directly modulate local immune and regenerative processes. Cytokines therefore offer several engineering advantages, including controllable dosing, relatively defined molecular targets, and compatibility with local delivery systems [194]. Their biological effects, however, are strongly dependent on concentration and exposure duration, and sustained presentation of the same cytokine at different stages of repair may produce distinct or even opposing outcomes. Accordingly, the objective of cytokine based design in immunomodulatory bone bioprinting should not be to maintain a particular macrophage phenotype continuously. Instead, material binding, local sustained release, and spatial patterning should be used to coordinate inflammatory regulation, cell recruitment, angiogenesis, and osteogenic signalling with the progression of bone repair [18,195].

IL-4 is one of the most extensively investigated immunoregulatory cytokines in osteoimmunomodulatory biomaterials, but its beneficial effects on bone repair cannot be attributed simply to an increase in M2 like macrophages. Zhao et al. incorporated IL-4 into three dimensional printed titanium scaffolds and demonstrated a clear dose and temporal dependence. Administration of 30 ng/mL IL-4 beginning on day 3 more effectively promoted a repair associated macrophage state and BMSC osteogenesis than either 100 ng/mL IL-4 or untreated conditions, indicating that the immunomodulatory outcome depends critically on both the timing and magnitude of exposure [144,195]. Zhang et al. further developed an IL-4 loaded calcium enriched gellan gum hydrogel and found that IL-4 treated macrophages exhibited increased TGF-β₁ expression, which subsequently activated TGF-β₁/Smad signalling in BMSCs, reduced apoptosis, and promoted osteogenic differentiation [196]. This finding establishes a more specific mechanistic link between changes in macrophage state and osteogenic signalling in MSCs and demonstrates that macrophage phenotypic markers alone are insufficient to explain bone formation. Wang et al. combined a polysaccharide hydrogel capable of co delivering IL-4 and BMP-2 with three dimensional printed titanium alloy scaffolds, thereby integrating immune regulation and direct osteogenic induction within the same system. Bone defect repair was superior to that achieved by scaffolds delivering either factor alone [197].

Local immobilisation and sustained release of cytokines can further enhance the capacity of printed scaffolds to regulate the immune microenvironment. Liu et al. constructed a GO/BP two dimensional heterostructured nanointerface on the surface of a three dimensional bioprinted polymer scaffold and exploited its large surface area to improve IL-4 loading and local release (Fig. 8A) [18]. IL-4 release increased the abundance of CD206 positive macrophages around the scaffold, while GO enhanced cell adhesion and BP continuously released phosphate species that supported osteogenesis. The regenerative effect of this system therefore resulted from the combined contributions of immunomodulation and intrinsic material osteoactivity rather than from macrophage phenotypic changes alone. In vitro, the construct enhanced both vascular responses of HUVECs and osteogenic activity of rBMSCs (Fig. 8B), while implantation in rat calvarial defects increased ALP and CD31 related signals together with new bone formation (Fig. 8C) [18]. A strategy that more closely reflects physiological bone repair is to regulate cytokine signals sequentially across different stages rather than continuously suppressing proinflammatory responses from the onset of implantation. Spiller et al. used short term IFN-γ delivery followed by sustained IL-4 release to regulate dynamic macrophage responses. Early proinflammatory macrophages favoured the initiation of vascular sprouting, whereas later repair associated macrophages supported vessel maturation, resulting in greater vascularisation with sequential delivery than with either cytokine alone [198]. This temporal principle has subsequently been incorporated into true bioprinting systems. Ma et al. incorporated IFN-γ loaded Laponite into a GelMA/alginate/four arm PEG acrylate bioink and fabricated microfibrous scaffolds using coaxial bioprinting and sequential crosslinking. The construct initially released IFN-γ to promote an early inflammation associated macrophage state. Subsequent degradation of Laponite released Mg²⁺ and Si related ionic species, gradually shifting macrophages towards a repair associated state. This transition was accompanied by increased endothelial progenitor cell migration, vascular sprouting, and vessel maturation, ultimately improving vascularized bone regeneration in calvarial defects [17]. Cytokines that primarily regulate cell recruitment, angiogenesis, or osteogenesis can also contribute to the sequential coordination of immune, vascular, and bone forming processes. SDF-1α primarily promotes recruitment of endogenous stem cells through CXCL12/CXCR4 mediated chemotaxis, and its material based delivery can simultaneously alter cellular recruitment and inflammatory responses at the implantation site [199,200]. In printed systems, multifactor combinations further illustrate the importance of spatial and temporal control. Bao et al. integrated VEGF165 and BMP-2 loaded hydrogels into FDM fabricated HA/PCL scaffolds, where sustained delivery supported angiogenesis and osteogenesis, respectively, and produced superior repair compared with single factor treatment in a rabbit segmental load bearing tibial defect model [19]. Cho et al. spatially patterned BMP-2 and FGF-2 within distinct collagen domains of three dimensional printed PCL/nHA scaffolds. Despite comparable total doses of the two factors, a radial distribution produced greater angiogenesis and endogenous bone formation, demonstrating that the value of printing lies not only in controlling total release but also in organising biological signals spatially [201].

Fig. 8.

A, Schematic illustration of osteoimmunomodulation mediated by a three-dimensional (3D) bioprinted scaffold functionalised with a two-dimensional (2D) heteronanostructure and the cytokine interleukin-4 (IL-4) to promote bone repair. B, Immunofluorescence results showing IL-4 released from the immunofunctionalized 3D scaffold, followed by induction of human umbilical vein endothelial cell (HUVEC) vascularisation and rat bone marrow mesenchymal stem cell (rBMSC) osteogenesis. C, Immunofluorescence staining of alkaline phosphatase (ALP) and CD31 proteins in bone defects and in defects implanted with the 3D bioprinted scaffold [18]. Copyright 2023, Elsevier.

Overall, cytokine functionalisation strategies are progressing from continuous delivery of single factors towards coordinated control of dose, timing, and spatial distribution. The effects of immunoregulatory cytokines such as IL-4 should therefore be evaluated in relation to intervention timing, local concentration, and downstream osteoimmune pathways such as TGF-β/Smad signalling, rather than interpreting increased CD206 expression alone as sufficient mechanistic evidence for bone regeneration. Similarly, a moderate early proinflammatory response retains physiological importance for tissue clearance, cell recruitment, and initiation of vascularisation. Bioprinting further enables multimaterial compartmentalisation, combination of distinct delivery vehicles, and cell co printing, providing the means to organise immunoregulatory cytokines, chemokines, angiogenic factors, and osteogenic factors according to different stages of repair. Future studies should more precisely define the effective dose window, receptor engagement, downstream signalling, local release kinetics, and interactions of individual factors, while minimising risks such as ectopic ossification associated with high dose BMPs, aberrant vascularisation, and prolonged immune deviation.

5.4.3. Vesicle-mediated immunoregulation: exosomes

Exosomes are extracellular vesicles that play important roles in intercellular communication. They can carry miRNAs, mRNAs, proteins, lipids, and other bioactive molecules and alter the transcriptional, metabolic, and differentiation states of recipient cells after cellular uptake (Fig. 9A and B) [[202], [203], [204], [205]]. Direct transplantation of living cells requires long term survival of implanted cells within the defect, whereas exosome function does not depend on such persistence and may reduce risks associated with abnormal cell proliferation and unintended differentiation. Exosomes are therefore particularly suitable for integration with printable hydrogels and porous scaffolds to develop cell free regenerative systems. However, exosomes are not homogeneous components with fixed anti inflammatory or osteogenic functions. Their biological effects are jointly determined by the parental cell type, activation state, preconditioning strategy, and molecular cargo [21]. Previous studies have shown that exosomes derived from hypoxia preconditioned MSCs can enhance angiogenesis and fracture repair through enrichment of miR-126 [204], demonstrating that exosome function can be modulated by altering the state of the parental cells. Accordingly, in immunomodulatory bone bioprinting, exosome source and functional state should be considered as important as the delivery strategy itself.

Fig. 9.

Exosome-based strategies for immunomodulation and bone regeneration. A, Schematic illustration of exosomes and their roles in bone regeneration. mRNA, messenger ribonucleic acid; miRNA, microRNA; lncRNA, long non-coding RNA; SM, phosphatidylinositol; PA, phosphatidylcholine [205]. Copyright 2022, Elsevier. B, Schematic illustration showing that exosome-loaded GNP hydrogel (GNP-Exos) promotes bone healing by immunomodulating macrophage polarisation. C, Effects of a bioprinted scaffold loaded with adipose-derived mesenchymal stem cell-derived exosomes (ADSC-Exos) on M1 macrophage polarisation in rats, assessed by immunofluorescence staining for inducible nitric oxide synthase (iNOS). D, Effects of the bioprinted scaffold loaded with ADSC-Exos on rat bone defects, evaluated by histological staining with hematoxylin and eosin (H&E) [203]. Copyright 2022, Springer Nature.

The regulation of the osteoimmune microenvironment by MSC derived exosomes is largely mediated by specific molecular cargo acting on macrophages, and their osteogenic effects should not be inferred solely from changes in M1 or M2 markers. Li et al. found that miR-451a enriched in ADSC derived exosomes targeted macrophage migration inhibitory factor, MIF, reduced inflammation associated markers such as iNOS, and increased CD206 expression. Incorporation of ADSC exosomes into a hydrogel containing gelatin nanoparticles enabled local release for approximately one week or longer and promoted new bone formation in a rat calvarial defect model (Fig. 9C and D) [203]. This study established a relatively clear molecular link among exosomal miRNA cargo, macrophage regulation, and bone repair. However, the delivery system was an injectable hydrogel rather than a three dimensional bioprinted construct and is therefore better regarded as mechanistic evidence supporting exosome mediated immunomodulation and local delivery.

Sun et al. more directly integrated immune cell derived exosomes with three dimensional printing and further demonstrated the source dependence of exosomal function [38]. The investigators first used β-TCP to modulate macrophage state and exosome secretion, after which the resulting bioceramic induced macrophage exosomes were incorporated into an alginate and hyaluronic acid formulation for extrusion based three dimensional bioprinting. β-TCP stimulation markedly altered the miRNA profile of macrophage derived exosomes, with differential expression of miRNAs associated with inflammatory regulation, osteogenesis, and angiogenesis involving signalling networks including TLR/TRAF6/NF-κB, MAPK, Wnt, and calcium signalling. Sustained exosome release enhanced BMP2, BMPR1A, Smad1, and Smad5 related signalling in BMSCs and improved angiogenic responses in HUVECs. Importantly, exosomes derived from macrophages exhibiting mixed functional states exerted stronger osteogenic effects on BMSCs than exosomes derived exclusively from M2 like macrophages, whereas their effects on several angiogenic outcomes were comparable [38]. Other studies further support this non-linear relationship between immune phenotype and osteogenesis. Luo et al. showed that miR-21a-5p enriched in exosomes derived from M1 like macrophages could enter osteoblasts and directly target GATA2, thereby relieving GATA2 mediated suppression of osteogenic differentiation, increasing expression of RUNX2, BMP2, and OCN, and promoting repair of rat calvarial defects [206]. These findings indicate that M1 associated signals are not necessarily antagonistic to bone formation and that exosomes derived from M2 associated macrophages are not uniformly more active across all stages of repair or osteogenic outcomes. Evaluation of exosome mediated osteoimmune regulation should therefore consider the state of the parental cells, exosomal miRNA and protein cargo, recipient cell type, and downstream pathways such as BMP/Smad, Wnt/β-catenin, and NF-κB, rather than constructing a single causal relationship based only on changes in iNOS and CD206.

Bioprinting further allows local retention, sustained release, and spatial colocalisation of exosomes with specific cell populations to become controllable design variables. Wang et al. incorporated Schwann cell derived exosomes together with BMSCs into a GelMA/SilMA bioink and fabricated SC-exos@G/S scaffolds by three dimensional bioprinting [20]. Incorporation of SilMA prolonged exosome release, while let-7c-5p carried by Schwann cell derived exosomes promoted BMSC osteogenic differentiation by targeting Acvr1b and regulating TGF-β signalling. The exosomes also enhanced endothelial progenitor cell migration and tube formation, leading to coordinated neural, vascular, and bone regeneration in a calvarial defect model. Chen et al. developed a bioink containing GelMA, decellularised bone matrix, and urine derived stem cell exosomes and fabricated GelMA/DBM/USC-Exos scaffolds using three dimensional bioprinting [21]. Exosomal activity was retained after printing, with sustained release extending beyond 16 days. The system reduced macrophage inflammatory responses associated with TNF-α, IL-6, and IL-1β while increasing IL-10 expression and repair associated phenotypes. These immunomodulatory effects were associated with changes in SOCS1/STAT3 and NF-κB related signalling. In mesenchymal stem cells, the exosomes further enhanced oxidative phosphorylation and osteogenic differentiation through AMPK related metabolic regulation. In vivo mandibular defect experiments showed simultaneous improvements in angiogenesis and new bone formation.

These studies indicate that the value of bioprinting for exosome based strategies extends beyond prolonging delivery. It enables exosomes, cells, and tissue specific matrices to be organised in predefined spatial relationships and thereby supports coordinated regulation of immune, vascular, and osteogenic processes. Exosomes provide a multicomponent regulatory strategy positioned between single cytokine delivery and living cell therapy, but this biological complexity also increases the difficulty of quality control. Future studies should first clearly define exosome source, parental cell culture conditions, and preconditioning protocols and establish standardised procedures for isolation, purification, particle size and concentration characterisation, and potency assessment. For bioprinted systems, it is also necessary to determine whether printing induced shear stress, photocrosslinking, or material components alter exosome membrane integrity or cargo activity, while reporting actual loading amounts, release kinetics, and in vivo retention. Mechanistic evaluation should move beyond simple assessment of M1 and M2 markers towards validation of specific exosomal cargo, target genes, and osteoimmune signalling pathways. Exosome depletion, inhibition of key miRNAs, or blockade of recipient signalling pathways should also be used to distinguish the independent contributions of the material and the exosomes. Only when exosome source, dose, delivery, and mechanism of action can be reproduced reliably will exosome functionalised bioprinted systems be positioned to advance towards standardisation and clinical translation.

5.4.4. Matrix-mediated immunoregulation and microenvironmental mimicry: dECM

Decellularised extracellular matrix (dECM) is produced by removing cellular components and major immunogenic constituents from native tissues while preserving, as far as possible, collagen, glycosaminoglycans, adhesion proteins, matrix bound growth factors, and tissue specific biochemical cues. It therefore provides cells with a microenvironment that more closely resembles native tissue than a single polymer network [207,208]. Pati et al. provided early evidence that tissue derived dECM could be solubilised, reformulated, and used as a cell laden bioink for three dimensional printing while retaining tissue specific effects on cellular behaviour [209]. In bone tissue engineering, Lee et al. subsequently combined bone derived dECM with alginate to formulate a cell laden bioink [210], while Dubey et al. incorporated amorphous magnesium phosphate into an ECM based formulation for craniofacial bone bioprinting [156]. These studies established the feasibility of dECM as a matrix component for bone bioinks. However, their principal endpoints focused on printability, cell viability, and osteogenic differentiation, and therefore do not provide sufficient evidence to infer a defined immunomodulatory function.

The immune effects of dECM arise from the collective interactions of multiple matrix components and degradation products with host cells rather than from a single defined receptor target, and these effects are strongly dependent on tissue origin. dECM derived from different tissues can elicit distinct macrophage responses. Comparative studies have shown that ECM derived from small intestinal submucosa, urinary bladder, oesophagus, brain, and colon can promote greater expression of repair associated macrophage markers, whereas skin derived ECM tends to induce a more pronounced proinflammatory phenotype. ECM derived from liver and skeletal muscle appears to exert more limited effects [211]. dECM therefore cannot be uniformly classified as a material that promotes an M2 phenotype. Its immune effects depend on tissue source, retained matrix composition, the extent of decellularisation, and subsequent enzymatic digestion and crosslinking. Residual DNA and cell membrane antigens may preserve immunogenicity, whereas matrix proteins, glycosaminoglycans, and bioactive degradation fragments retained after adequate decellularisation may influence macrophage adhesion, migration, and inflammatory mediator expression. Evaluation of the osteoimmune effects of dECM should therefore consider both matrix composition and host cell responses rather than relying solely on CD206 or other individual phenotypic markers.

This principle was directly examined in the three dimensional bioprinted periodontal regeneration system reported by Yang et al. [22]. Dental follicle derived dECM was incorporated into GelMA to formulate a GelMA/dECM bioink containing dental follicle stem cells, and DLP and extrusion based bioprinting were used to construct periodontal ligament and alveolar bone functional modules, respectively (Fig. 10A). The formulation containing 5 mg/mL dECM maintained favourable printing performance while enhancing cell migration and post printing survival, together with increased expression of osteogenic markers including RUNX2, ALP, and OCN (Fig. 10B). More importantly, in LPS stimulated RAW264.7 macrophages, dECM markedly reduced expression of proinflammatory genes including IL 1β, IL 6, TNF α, and MCP1. Some of these markers were reduced by approximately 50% relative to the LPS group, and secretion of IL 6 and TNF α proteins was also decreased. Following subcutaneous implantation in rats, GelMA/dECM constructs showed less inflammatory cell infiltration and fibrous encapsulation than GelMA alone, together with a lower proportion of proinflammatory macrophages. In a subsequent canine periodontal defect model, the construct promoted alveolar bone formation, collagen fibre maturation, and reconstruction of the bone periodontal ligament interface (Fig. 10C) [22]. Importantly, although this study demonstrated that dECM can attenuate excessive inflammation and support tissue regeneration, the specific matrix components and molecular mechanisms responsible for these immunomodulatory effects remain incompletely defined. A more accurate interpretation is therefore that dECM can reshape the local inflammatory environment and support subsequent osteogenesis, rather than acting through a single macrophage polarisation mechanism.

Fig. 10.

A, Schematic illustration of three-dimensional (3D) bioprinting for periodontal regeneration using a gelatin methacryloyl/decellularized extracellular matrix (GelMA/dECM) bioink scaffold. B, In vitro results showing that the 3D bioprinted dECM scaffold promoted osteoblastic differentiation, as assessed by immunofluorescence staining of osteogenic markers, including alkaline phosphatase (ALP), runt-related transcription factor 2 (Runx2), and osteocalcin (OCN). C, Sirius Red staining under polarized light and Movat–Russell staining demonstrating that the bioprinted dECM scaffold promoted bone regeneration and increased collagen fiber maturation [22]. Copyright 2022, Wiley. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Recent bone specific bioprinting studies have further expanded the role of dECM from passive matrix support towards stage dependent immune regulation. Liu et al. methacrylated bone derived dECM to generate bdECM MA and combined it with silicon substituted calcium phosphate and BMSCs for DLP bioprinting of tissue engineered bone constructs [11]. The system maintained high cell viability, achieved mechanical properties at the MPa scale, and exhibited osteogenic, angiogenic, and immunomodulatory activity. More importantly, the in vivo immune response was characterised by early activation of proinflammatory responses followed by a gradual transition towards anti inflammatory and tissue reparative states, rather than continuous maintenance of a single repair associated macrophage phenotype from the onset of implantation. Mechanistic analysis suggested that subsequent immune regulation was associated with inhibition of p38 MAPK signalling. These findings indicate that bioinks combining bone derived dECM with inorganic phases can participate in the temporal regulation of inflammation and thereby influence subsequent vascularisation and osteogenesis. They also further support the concept that bone regeneration requires stage appropriate immune transitions rather than sustained elevation of M2 associated markers.

A further advantage of dECM lies in its capacity to provide spatially regionalised regenerative signals through tissue specific matrix composition. Zhang et al. developed a three dimensional bioprinted scaffold with structural, compositional, and mechanical gradients and functionalised different regions with tendon, cartilage, and bone derived dECM. This design generated gradual regional differences in tenogenic, chondrogenic, and osteogenic induction and improved both histological and mechanical reconstruction of the tendon to bone interface in a rabbit rotator cuff repair model [212]. Kim et al. further used bone and tendon specific dECM bioinks together with gradient core shell printing to construct a multilayered tendon to bone interface that directed stem cell differentiation towards regionally appropriate lineages [213]. These studies did not directly establish dECM mediated osteoimmune mechanisms and therefore should not be regarded as direct evidence of macrophage regulation. Nevertheless, they demonstrate that bioprinting can exploit the tissue specificity of different dECM sources to establish spatially organised matrix signals within a single construct. For complex bone defects, this spatial organisation could in future be integrated with immunomodulatory, angiogenic, and osteogenic signals.

Overall, the principal value of dECM in immunomodulatory bone bioprinting lies in its capacity to provide tissue specific matrix composition, a supportive cell adhesion environment, and degradable biological signals, while bioprinting enables these features to be organised spatially. This complexity, however, also creates substantial challenges for standardisation. Donor characteristics, tissue location, and age can alter dECM composition, while decellularisation, enzymatic digestion, sterilisation, and chemical crosslinking may damage glycosaminoglycans, matrix proteins, and matrix bound growth factors and consequently modify host immune responses. dECM also generally lacks sufficient rheological stability and mechanical strength for independent printing and therefore often requires combination with GelMA, alginate, inorganic particles, or chemical modification to improve printability and structural retention [214]. Future studies should therefore extend beyond measurements of osteogenic and macrophage markers and more clearly define dECM source, residual DNA content, key matrix components, degradation products, post printing biochemical activity, and the corresponding immune signalling pathways. Component dissection and pathway inhibition should also be used to distinguish the independent contributions of dECM from those of inorganic phases, crosslinked networks, and encapsulated cells. Only when tissue source, processing procedures, and immune effects can be standardised will dECM be positioned to progress from a tissue mimetic bioink towards a reproducible immunomodulatory material for bone regeneration.

5.5. Spatiotemporal and intelligent immunomodulatory strategies

Material composition, structural features, and immunologically active components provide multiple effective approaches for regulating the immune microenvironment of bone defects. However, most existing systems still operate through relatively static regulatory modes, in which material properties and bioactive signals are largely predetermined before implantation and subsequently act on host tissues according to predefined patterns [215]. The immune requirements of bone repair, however, change continuously over time. During the early stage of injury, neutrophils, monocytes, and inflammation associated macrophages contribute to pathogen clearance, removal of necrotic tissue, and initiation of repair. The immune response must subsequently transition towards inflammatory resolution and further coordinate with T cells, B cells, vascular cells, MSCs, and osteoclasts to support bone formation and remodeling [216]. Continuous suppression of inflammation or prolonged delivery of a single pro-reparative signal may therefore be inconsistent with the physiological sequence of bone healing. Spatiotemporal regulation can improve the temporal precision of therapeutic intervention through stage specific release and microenvironment responsive mechanisms, while AI assisted design, four dimensional bioprinting, and digital manufacturing may further enhance the predictability and adaptability of material design and fabrication.

5.5.1. Spatiotemporal and responsive immunomodulation

The central aim of spatiotemporal immunomodulation is not simply to prolong the release of bioactive agents, but to align material function with the changing immune requirements of different stages of repair. Programmed regulation typically relies on differences in degradation rate, binding affinity, or spatial compartmentalisation to enable multiple signals to be released in a predefined sequence. Responsive regulation further exploits local pathological cues within bone defects, including pH, ROS, enzyme activity, and other microenvironmental signals, to alter material degradation, crosslinking, or drug release, thereby allowing the intensity of intervention to change with the local tissue state [217,218]. A more advanced strategy should therefore integrate predefined repair sequences with local pathological signals rather than treating programmed and responsive regulation as mutually exclusive approaches.

This concept is particularly relevant to infected bone defects, where premature suppression of inflammation may compromise pathogen clearance, whereas persistent inflammation can impair subsequent vascularisation and osteogenesis. Jin et al. developed a time dependent immunomodulatory hydrogel that exploited distinct ion release kinetics, with early Ca²⁺ release maintaining an inflammatory response favourable for infection control, followed by sustained Zn²⁺ release to promote inflammatory resolution and osteogenesis. This design avoided reducing immunomodulation to continuous promotion of an M2 like macrophage phenotype [219]. Its significance lies in matching release kinetics to pathological stages rather than simply increasing the loading of a particular anti inflammatory factor. Zhao et al. further integrated a pH responsive hydrogel and metal polyphenol nanoparticles into a three dimensional printed HA scaffold. Early signals first remodeled the macrophage associated inflammatory environment, after which the acidic microenvironment triggered the release of Sr²⁺ and other components, further enhancing the immunomodulatory function of recruited MSCs. Single cell analysis showed that the system not only altered macrophage states but also induced an MSC subpopulation with immunoregulatory characteristics, involving CCL2/CCR2 mediated cellular interactions [220]. Spatiotemporal design is therefore progressing from modulation of a single immune cell phenotype towards regulation of continuous osteoimmune cascades involving immune, stromal, and osteogenic cell populations.

Similar multistage strategies are beginning to extend to signals associated with adaptive immunity. Sui et al. engineered a combination of regulatory T cell derived exosomes and the BMP-2 mimetic peptide P24, in which Treg derived exosomes provided early immunomodulatory effects, followed by a more direct osteogenic stimulus from P24, thereby achieving stage specific coupling between immune regulation and osteogenic induction [221]. Although this study did not employ a bioprinting platform, its design principle is highly relevant to bioprinting because multimaterial printing can further control the spatial localisation and local dose of different vesicles, cytokines, or cell populations within a scaffold. More broadly, future spatiotemporal regulation should not remain restricted to macrophages. Depending on the stage of bone repair, relevant targets may include early neutrophil clearance and NET control, DC maturation and antigen presentation, Treg/Th17 balance, B cell mediated regulation of the RANKL/OPG axis, and osteoclast activity during later remodeling. Direct studies of these immune populations in responsive bone biomaterials remain substantially less developed than those focusing on macrophages, representing an important gap for future research.

Microenvironment responsive systems can further improve state dependent intervention beyond predefined sequential release. Xie et al. developed an acid responsive hydrogel for infected bone defects that was designed around the dynamic features of the pathological environment, including early oxidative stress, local acidosis, and persistent infection. The system accelerated the release of functional carbon dots under pathological conditions and subsequently continued to regulate oxidative stress, infection control, bone formation, and osteoclast balance [222]. Fang et al. further exploited elevated nitric oxide levels within injured bone as a biological trigger and engineered a living hydrogel capable of sensing NO and responding by secreting BMP-2, thereby shifting the strategy from passive drug release towards a sense and respond mode of regulation [223]. Although the principal output of this latter system was not directed specifically towards immune cells, the sensing and response paradigm provides a methodological basis for developing future bioinks capable of detecting inflammatory intensity and adjusting the release of immunologically active signals accordingly. Future responsive materials should therefore progress beyond relatively nonspecific inflammatory cues such as pH and ROS towards signals that more directly reflect specific immune states. Quantitative relationships among stimulus intensity, release dose, and immune cell responses will also need to be established, as greater mechanistic complexity does not necessarily translate into more reliable bone regeneration.

5.5.2. AI-assisted optimisation of bioinks and scaffolds

Immunomodulatory bioinks inherently represent a high-dimensional, multi-objective design problem. Material type and concentration determine not only rheological properties, crosslinking behaviour, and printing fidelity, but also degradation kinetics, release of ions or bioactive agents, cell survival, and host immune responses. Conventional experimental approaches that vary one parameter at a time are poorly suited to resolving the nonlinear relationships among these variables, whereas machine learning can establish mappings between material, rheological, structural, and biological inputs and corresponding printing or functional outcomes, thereby reducing empirical trial and error [224]. Existing studies have demonstrated that machine learning can predict bioink printability from parameters such as elastic modulus and yield stress and can further identify key rheological features governing hydrogel printability [225,226]. Although these studies have primarily addressed manufacturing performance, they provide a foundation for incorporating immunomodulatory outcomes into future bioink optimisation frameworks.

At the scaffold level, the main value of AI lies in coordinated optimisation across multiple design variables. Ibrahimi et al. developed machine learning models using more than one thousand TPMS architectures, enabling scaffold design to simultaneously account for mechanical, morphological, and manufacturing constraints [227]. Lin et al. further applied machine learning to analyse a database of osteoinductive biomaterials and identified variables such as porosity, HA content, and BMP-2 as important determinants of osteoinductive outcomes, after which the model was used to guide optimisation of material and structural parameters [228]. At present, however, the optimisation endpoints of most such models remain focused on printability, mechanical performance, and osteogenic outcomes, while immune responses are rarely included as routine outputs. This limitation also highlights the largely unrealised potential of AI in immunomodulatory bioprinting. Future models could incorporate material composition, ion release, degradation kinetics, pore architecture, and printing parameters as inputs, while treating macrophage functional states, neutrophil activation, DC maturation, Treg/Th17 balance, inflammatory cytokine profiles, osteoclastogenesis, osteogenesis, and vascularisation as multi-objective outputs. Such an approach would extend AI optimisation beyond the generation of printable constructs towards the design of scaffolds that are better matched to the desired temporal sequence of immune responses.

This direction is highly complementary to advances in single cell transcriptomics, spatial transcriptomics, and high-throughput biomaterial screening. High-dimensional omics can reveal immune cell subsets and intercellular communication patterns that cannot be captured by a small number of conventional markers such as CD86 and CD206, thereby providing more biologically informative labels for model training. Recent machine learning studies in biomaterials are also moving beyond isolated prediction of material properties towards multiscale design frameworks that integrate composition, processing, and biological responses [229]. However, several limitations remain. Bioprinting datasets are generally small, while material formulations, printing parameters, cell sources, and immune evaluation criteria vary substantially among laboratories. Bone regeneration outcomes also often require weeks or months to obtain, and differences in animal species and pathological models further limit model generalisability. AI-assisted immunomodulatory bioprinting therefore requires interpretable models, uncertainty quantification, independent validation, and standardised immune and bone regeneration endpoints rather than relying solely on predictive accuracy.

5.5.3. 4D bioprinting and digital intelligent manufacturing

Four-dimensional (4D) bioprinting extends conventional three-dimensional bioprinting by incorporating a temporal dimension, enabling printed constructs to undergo changes in shape, architecture, mechanical properties, or biological function in response to external or endogenous stimuli [230]. In the context of osteoimmunomodulation, its potential extends beyond simple shape recovery and lies in the possibility of dynamically altering pore architecture, stiffness, interfacial properties, and bioactive factor release as repair progresses [231,232]. During the early stage, an open porous architecture and enhanced mass transport may facilitate inflammatory cell infiltration and tissue clearance, whereas subsequent changes in structure or mechanical state may further support vascularisation and osteogenesis. Stimuli-responsive 4D systems can also be integrated with temporally controlled delivery of drugs or cytokines, allowing structural changes to occur in parallel with changes in immune signalling [233]. Current 4D bone scaffold research, however, remains focused predominantly on shape memory, morphological transformation, and drug delivery, with temperature-responsive materials accounting for a substantial proportion of reported systems. Direct studies targeting neutrophils, DCs, T and B cells, or multicellular osteoimmune networks remain very limited [234,235]. It is therefore more appropriate at present to regard 4D bioprinting as a promising technological platform for dynamic osteoimmune regulation rather than as an established immunomodulatory strategy. Digital twins and intelligent manufacturing, by contrast, primarily address the predictability and manufacturing reproducibility of such complex systems. A digital twin creates a virtual model that remains continuously linked to the physical printing process and can integrate patient imaging, material parameters, printing status, and subsequent performance data, while using model prediction and feedback to adjust manufacturing parameters. Ahmed et al. have already established a digital twin framework for three-dimensional bioplotting of PCL scaffolds to integrate process data and optimise printing quality and subsequent performance [236]. At the same time, real-time monitoring is gradually transforming bioprinting systems from devices that simply execute predefined toolpaths into platforms capable of simultaneously tracking bioink pH, temperature, viscosity, cell concentration, cell viability, and structural defects, thereby providing the data required for closed-loop feedback control [237].

For immunomodulatory bioprinting, future incorporation of actual bioactive factor release profiles, cellular states, and local immune biomarkers into digital models could potentially enable continuous prediction from patient-specific defect characteristics to bioink formulation, structural design, manufacturing quality, and the anticipated immune response. At present, however, direct application of digital twins to osteoimmunomodulatory bioprinting remains largely conceptual or at an early validation stage. Further development will depend on reliable sensing technologies, harmonised data standards, biologically validated predictive models, and quality control frameworks compatible with clinical manufacturing requirements.

5.6. Pathology-specific osteoimmune microenvironments and implications for bioprinting

Most in vivo evidence for immunomodulatory bone repair materials still derives from standard bone defect models in young, healthy animals. In these models, post-injury inflammation generally progresses towards resolution and tissue repair, making it relatively straightforward to link material-induced changes in macrophage state with subsequent osteogenesis. Clinically complex bone defects are fundamentally different. Infection, metabolic disorders, ageing, and other pathological conditions may disrupt local immune homeostasis even before injury occurs and further alter the duration and trajectory of immune responses during bone repair [238]. Consequently, reductions in M1-associated signals or increases in M2-associated markers observed in healthy animals should not be regarded as universal design goals for all pathological bone defects. For bioprinting, a more important consideration is to identify the dominant immune abnormality within a specific pathological microenvironment and to adjust the timing and mode of material intervention accordingly.

Both infected and diabetic bone defects are characterised by persistent inflammation, but their regulatory requirements are distinct. In infected environments, bacteria and biofilms continuously activate innate immunity, meaning that an early proinflammatory response remains necessary for pathogen clearance. Recent studies suggest that infected bone repair may benefit more from stage-specific immunomodulation, in which sufficient antimicrobial immunity is maintained during the infection-control phase before subsequent promotion of inflammatory resolution and bone formation [219,239]. Studies of three-dimensional printed scaffolds for infected bone defects have likewise begun to integrate antibacterial activity with subsequent macrophage regulation rather than simply pursuing sustained anti-inflammatory effects after implantation [240]. Persistent inflammation in diabetic bone defects is more strongly driven by hyperglycaemia and oxidative stress and is accompanied by vascular dysfunction and abnormal bone remodeling [241]. Fu et al. developed a glucose- and ROS-responsive system that simultaneously improved the metabolic microenvironment and immune response and subsequently restored vascular and bone remodeling, indicating that diabetic bone repair requires intervention against persistent pathological stimuli [242]. Recent three-dimensional printing studies have also shown that local metformin delivery or the use of tissue-derived matrices can improve both immune regulation and bone regeneration in diabetic defects [243]. Notably, Sun et al. previously used true multicellular three-dimensional bioprinting to integrate BMSCs, macrophages, and BMP-4 delivery within a single construct and improved both dysregulated inflammation and bone formation in calvarial defects in diabetic rats [150]. Collectively, these findings indicate that infected defects require effective pathogen clearance before transition towards repair, whereas diabetic defects require sustained correction of the abnormal metabolic microenvironment.

Ageing, osteoporosis, and radiation-induced bone injury share a reduction in intrinsic regenerative capacity, although their underlying immune abnormalities differ [244]. Single-cell studies have shown marked alterations in macrophage subsets and transcriptional states within aged fracture tissue, accompanied by dysregulation of TREM2-related inflammatory programmes [245]. Building on this concept, Sun et al. incorporated Mg/Ce-based metal-organic frameworks into a cell-laden bioink and promoted bone defect repair in aged rats by reducing oxidative stress and improving the senescent microenvironment [246]. These findings suggest that bone regeneration in ageing cannot rely on conventional osteogenic stimulation alone but also requires correction of immune dysfunction associated with cellular senescence. Osteoporotic defects are characterised not only by inflammatory abnormalities but also by pronounced imbalance in bone turnover. Recent three-dimensional printed scaffolds have therefore begun to combine early immune modulation with subsequent control of bone metabolism. Examples include sequential release of IL-4 and alendronate to promote bone formation while limiting excessive osteoclast activity [247], as well as strategies that improve vascularisation to restore local bone metabolism under osteoporotic conditions [248]. Evaluation of these models should therefore extend beyond macrophage markers to include long-term bone remodeling. Radiation-associated bone injury is further complicated by microvascular damage and persistent oxidative stress caused by irradiation. Wang et al. showed that an Mg²⁺-containing hydrogel alleviated radiation-induced bone injury by modulating macrophage responses and improving the microvascular environment [249], while low- and high-dose irradiation may exert different effects on macrophage and osteogenic responses [250]. These findings indicate that bone defects following radiotherapy require explicit consideration of radiation-induced alterations in the host microenvironment, although direct bioprinting evidence for this pathological setting remains limited.

Tumour resection-associated bone defects present an additional challenge because a pro-regenerative immune environment may conflict with tumour control. Signals that promote inflammatory resolution and angiogenesis during conventional bone repair may not necessarily be safe when residual tumour cells are present. Immunomodulatory objectives in tumour-associated defects should therefore change according to the treatment stage. Li et al. developed a three-dimensional printed MgO₂/PLGA scaffold that enhanced antitumour activity and promoted an M1-like macrophage response through early H₂O₂ release, followed by sustained Mg²⁺ release that favoured a repair-associated immune state and osteogenesis. This design exemplified a temporal transition from tumour control to bone regeneration [251]. Although this system involved three-dimensional printing rather than cell-laden bioprinting in the strict sense, its design principle remains highly informative. Overall, no universal immunomodulatory strategy is suitable for all pathological bone defects. Future bioprinting studies should therefore move beyond increasing material and structural complexity and make greater use of disease-specific animal models, while directly comparing the immune responses elicited by the same construct under healthy and pathological conditions to define its true range of applicability.

6. Clinical translation

The clinical translation of immunomodulatory bioprinted materials for bone repair depends not only on whether they can promote new bone formation in animal models, but also on whether these biological effects can be translated into clinical products with well defined compositions, robust manufacturing processes, and reproducible quality attributes. As immunologically active components such as stem cells, cytokines, exosomes, and dECM are incorporated into bioinks, manufacturing and quality assessment become substantially more complex than those required for conventional acellular printed scaffolds [252,.253]]. Cells and naturally derived materials from different sources may exhibit batch to batch heterogeneity, the effective dose and stability of bioactive components may vary, and printing parameters can further influence cell viability, construct architecture, and release behaviour. Clinical development therefore requires the early establishment of critical quality attributes that correspond to the intended function of the product, rather than relying solely on scaffold geometry and compressive strength as release criteria [254,255]. For immunomodulatory products, these attributes should also reflect the intended biological functions, including the identity and activity of living cells, the effective content and release characteristics of bioactive agents, and the potency of the product in regulating inflammatory responses. Accordingly, the central requirement of good manufacturing practice (GMP) is not simply to transfer a laboratory printing process into a cleanroom, but to integrate raw material sourcing, formulation preparation, printing parameters, culture procedures, and final product assessment into a validated and traceable manufacturing system.

Sterility control is a particularly important yet frequently underestimated component of this manufacturing framework. Many bone bioinks are based on naturally derived materials such as GelMA, collagen, hyaluronic acid, and dECM, and sterilisation procedures may alter polymer molecular structure, viscosity, crosslinking behaviour, and mechanical properties, thereby shifting the original printing window. Menon et al. systematically compared sterilisation approaches for different bioink components and showed that heat, ultraviolet irradiation, ethylene oxide, filtration, and other treatments exerted markedly different effects on the rheological properties, mechanical performance, and biocompatibility of individual materials [256]. Sterilisation should therefore not be treated as an additional step introduced after bioink development, but should instead be optimized in parallel with material formulation and printing conditions. Terminal sterilisation is particularly difficult for constructs containing living cells, making these products more dependent on sterile raw materials and aseptic manufacturing. For systems containing exosomes, cytokines, or dECM, it is also necessary to determine whether sterilisation alters molecular composition or immunological activity [3,4]. This issue is especially important for the immunomodulatory materials discussed in this review, because a processing step may leave scaffold morphology largely unchanged while substantially altering cytokine release or immune cell responses. Future quality assessment should therefore extend beyond conventional sterility, structural integrity, and cell viability tests to include product specific assays of immunomodulatory potency.

At the manufacturing level, bioprinting must also address the inherent tension between personalisation and standardisation. Patient specific reconstruction of defect geometry from medical imaging represents one of the major clinical advantages of bioprinting. However, if every construct differs completely in geometry, material composition, and printing parameters, robust process validation and batch consistency become difficult to establish. A more practical translational pathway may therefore involve standardising raw materials, bioink formulations, and manufacturing platforms while restricting patient specific modifications to a predefined and validated design space, thereby combining a standardised production platform with personalised geometric design [257]. Recent studies addressing the scale up of bone and cartilage bioprinting have similarly emphasized the transition from manual laboratory fabrication towards standardised bioinks, automated cell expansion, clearly defined lot release criteria, and continuous or parallel manufacturing [258]. In this context, scalability does not necessarily require a single machine to produce large numbers of identical scaffolds. For patient specific or cell laden products, a scale out strategy in which a standardised platform supports parallel production of multiple batches may be more appropriate, with automated process monitoring used to minimise variability arising from operators and manufacturing sites.

Ultimately, product complexity must be justified by a corresponding clinical benefit. Immunomodulatory bioprinting systems can incorporate increasing numbers of cell types, cytokines, responsive materials, and multistage release modules, but every additional component introduces new requirements for raw material control, process parameters, stability assessment, and quality testing and may substantially increase manufacturing failure rates and overall cost [255,259]. Clinical translation should therefore not necessarily pursue the most functionally complex bioprinted system, but rather identify the minimum level of complexity required to achieve a robust therapeutic effect. For some bone defects, acellular scaffolds combined with controlled ion release or exosome delivery may be more readily scalable than complex multicellular systems. For defects that genuinely require patient derived cells or highly personalised reconstruction of the local microenvironment, the additional manufacturing burden must be justified by a clinically meaningful improvement in therapeutic outcome. Artificial intelligence, automated printing, and in process quality monitoring may reduce the costs associated with formulation screening and process optimisation, but their practical value will ultimately depend on whether they can reduce batch failures, shorten production times, and improve product consistency [259]. The future clinical competitiveness of immunomodulatory bioprinted materials will therefore be determined not only by regenerative efficacy, but also by their ability to balance biological effectiveness, manufacturing complexity, regulatory controllability, and economic affordability.

7. Conclusions and perspectives

Bone regeneration is a dynamic osteoimmune process in which inflammation, vascularisation, progenitor cell recruitment, osteogenic differentiation, and bone remodeling are tightly coordinated across different stages of healing. Accordingly, the function of bone regenerative biomaterials should extend beyond structural support or direct osteogenic stimulation towards active regulation of the local immune microenvironment. Three-dimensional bioprinting provides a suitable platform for this purpose because material composition, architecture, cellular organisation, and bioactive signals can be spatially integrated within a single construct. As summarised in this review, natural and synthetic polymers, inorganic ions, structural and topographical cues, stem cells, cytokines, exosomes, and dECM can regulate immune responses through distinct but interconnected mechanisms, thereby influencing angiogenesis, osteogenesis, and subsequent tissue remodeling. Importantly, effective osteoimmunomodulation should not be reduced to persistent suppression of inflammation or continuous promotion of an M2-like macrophage phenotype. Instead, successful regeneration requires preservation of an appropriate early inflammatory response, timely transition towards inflammatory resolution and tissue reconstruction, and subsequent restoration of long-term immune and bone metabolic homeostasis.

Future studies should therefore move from static and single-target immunomodulation towards stage-specific and multicellular regulation. Current studies still rely heavily on macrophage-related markers, whereas the roles of neutrophils, osteomacs, dendritic cells, T cells, B cells, osteoclasts, and their interactions with vascular and stromal populations remain insufficiently characterised. Single-cell sequencing, spatial transcriptomics, and other high-resolution approaches may help identify stage-specific immune states and spatial cell interactions, providing a more mechanistic basis for biomaterial design. At the material level, programmable degradation, sequential delivery, spatial compartmentalisation, and microenvironment-responsive systems may enable immunoregulatory signals to better match the changing requirements of inflammatory initiation, proliferative repair, and remodeling. Multimaterial and 4D bioprinting may further allow dynamic control of structure, mechanics, permeability, and bioactive release after implantation. In parallel, artificial intelligence and data-driven optimisation may assist in integrating bioink composition, printing parameters, degradation behaviour, structural features, and biological responses, thereby improving the efficiency and precision of scaffold design.

Clinical translation, however, will require biological sophistication to be balanced against manufacturability, reproducibility, safety, and regulatory feasibility. The incorporation of living cells, cytokines, exosomes, dECM, and multifunctional delivery systems can enhance regenerative performance, but also increases variability in raw materials, manufacturing processes, storage, and quality control. Future development should therefore establish critical quality attributes that reflect not only scaffold geometry and mechanical properties, but also the stability, potency, and reproducibility of the intended immunoregulatory function. Standardised manufacturing, clinically relevant large-animal models, long-term evaluation of degradation and remodeling, and scalable GMP-compatible production will be essential. Overall, the next generation of immunoregulatory bioprinting should not simply incorporate more functional components, but should achieve more precise, stage-matched, and clinically controllable regulation of the osteoimmune microenvironment, ultimately enabling more reliable restoration of bone structure and function.

CRediT authorship contribution statement

Xiaonan Wang: Writing – review & editing, Data curation. Aobo Zhang: Writing – review & editing, Writing – original draft.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

No data was used for the research described in the article.