Abstract

Background

   Radiation-induced jaw injury is one of the most severe complications
   after radiotherapy for head and neck cancer, which can disrupt
   patients’ health and quality of life. Although the direct target of
   inflammation and suppressed bone regeneration activity by ionizing
   radiation (IR) has been phenomenally observed, the underlying
   mechanisms and potential therapeutic targets remain blurred.
   Osteoimmunology emphasizes that dendritic cells (DCs) may contribute to
   bone diseases.

Methods

   In this study, we assessed phenotypic and functional alterations of DCs
   in a radiation-induced jaw injury rat model through
   immunohistopathological staining. The effects of IR on bone
   marrow-derived dendritic cells (BMDCs) in vitro were further validated
   by flow cytometry, ELISA, mixed lymphocyte reaction (MLR) assay, and
   transwell. The cellular responses and differentiation of bone marrow
   mesenchymal stem cells (BMSCs) under BMDC-derived conditioned medium
   stimulation were evaluated through various cell staining, Quantitative
   real-time polymerase chain reaction (qRT-PCR), and western blotting.
   Flow cytometry, qRT-PCR, WB were employed to verify tolerogenic
   characteristics of Vitamin D3 (VitD3)-induced tolerogenic DCs (tolDCs).
   TolDCs were encapsulated in GelMA to develop an effective in vivo
   therapeutic approach for irradiated jaw defects.

Results

   We revealed that IR activated the mature-inflammatory phenotype and
   corresponding biological functions of DCs through the nuclear factor
   kappa-B (NF-κB) signaling pathway. Exposure to conditioned medium from
   irradiated BMDCs induced oxidative stress and inflammatory responses in
   BMSCs, inhibiting their proliferation, migration, and osteogenic
   potential while potentiating adipogenic capacity. Furthermore, tolDCs
   were proven to be resistant to radiation-induced activation. Local
   administration of tolDCs was effective in improving bone regeneration
   of irradiated jawbone defects.

Conclusions

   Hyperactivation of DCs served as a potential pathogenic factor in
   radiation-induced jaw injury, exacerbating local inflammation and
   abrogating the biological functions of BMSCs. The local transplantation
   of tolDCs was a promising therapeutic strategy for osteogenesis in
   radiation-induced jaw injury.

Supplementary Information

   The online version contains supplementary material available at
   10.1186/s13287-025-04508-x.

   Keywords: Osteonecrosis of the jaw, Lonizing radiation, Dendritic
   cells, Osteoimmunology, Bone marrow mesenchymal stem cells

Introduction

   Radiotherapy (RT) plays a pivotal role in managing head and neck
   cancer, with approximately 80% of the patients undergoing RT at least
   once during the course of their disease [[34]1]. Radiation-induced jaw
   injury is one of the potentially severe, delayed, and irreversible
   complications of RT, occurring in up to 30% of the patients, and the
   incidence of osteoradionecrosis of the jaw (ORNJ) is about 8–15%
   [[35]2–[36]4]. The characteristic clinical features of
   radiation-induced jaw injury include persistent pain, infection,
   exposure of necrotic bone, and ultimately functional injuries [[37]5].
   At present, the gold standard therapeutic approaches for patients with
   radiation-induced jaw injury have not yet been defined. Possible
   interventions start from non-invasive measures like oral hygiene and
   antibiotics, escalate through sequestrectomy to segmental
   mandibulectomy [[38]6]. The symptoms and surgical treatments have
   negative impacts on practically all patient-related quality of life
   metrics and remain a medical challenge to surmount [[39]7].
   Understanding the pathophysiology of radiation-induced jaw injury is
   crucial for developing effective and targeted treatment strategies,
   while there remain many gray areas to be explored at the cytological
   and molecular levels [[40]8–[41]10].

   With the development of osteoimmunology, growing evidences highlight
   the extensive crosstalk between the skeletal and immune systems
   [[42]11]. Our previous studies [[43]12, [44]13] have shown that
   ionizing radiation (IR) alters the composition, activity, and
   immunological status of immune cells in the jawbone cavity, thus
   breaking the homeostasis of the immune microenvironment at the single
   cell level, and suggested that there existed a correlation between
   dendritic cells (DCs) and early stage radiation-induced jaw injury.

   DCs are central components of innate immunity and regulators of
   adaptive immunity. DCs are not only actively involved in all phases of
   tissue healing, from initial inflammation to tissue remodeling
   [[45]14], but also play a regulatory role in skeletal homeostasis and
   bone regeneration [[46]15]. Depending on the external signals, immature
   DCs (imDCs) can be selectively differentiated into either a mature or
   tolerogenic state [[47]16]. A biased differentiation of DCs may result
   in exacerbated inflammation or loss of tolerance. For example, an
   increase in highly activated DCs was observed in the progression of
   osteonecrosis of the femoral head [[48]17, [49]18]. However, to date,
   there is a lack of in-depth investigation into the correlation between
   DC phenotypic/functional alterations and radiation-induced jaw injury.

   Bone marrow mesenchymal stem cells (BMSCs), capable of differentiating
   into osteoblastic lineages and promoting osseointegration, have emerged
   as a central focus and target cell in the field of regenerative
   medicine [[50]19, [51]20]. Despite several studies have demonstrated
   the regulatory role of immune cells on functions of BMSCs, there is no
   agreement on which DC phenotype is beneficial or detrimental for BMSCs
   [[52]15, [53]21]. A better understanding of the crosstalk between DCs
   and BMSCs is proposed to have a positive impact on tissue regeneration
   in radiation-induced jaw injury.

   Tolerogenic DCs (tolDCs) refer to a maturation-resistant DC subset
   induced by antigenic or pharmacological stimuli, characterized by their
   low expression of co-stimulatory molecules, high expression of
   tolerance-associated markers, and anti-inflammatory cytokine production
   [[54]22]. Multiple studies have reported beneficial effects of tolDC
   therapy in the treatment of non-self-limiting inflammatory diseases and
   chronic wounds [[55]23, [56]24]. We hypothesize that tolDCs may also
   have therapeutic effects on radiation-induced jaw injury.

   Here, we observed the recruitment of DCs with enhanced
   maturation-related phenotypic and functional characteristics in a
   radiation-induced jaw injury rodent model. In vitro experiments further
   demonstrated that IR promoted the transition of DCs toward a mature and
   inflammatory phenotype via the nuclear factor kappa-B (NF-κB) signaling
   pathway. Conditioned medium (CM) from these irradiated DCs induced
   oxidative stress, inhibited osteogenic, and promoted lipogenic
   differentiation of BMSCs. Vitamin D3 (VitD3)-treated tolDCs exhibited
   radiation-resistant features, which effectively mitigated mature DC
   (mDC)-mediated undesired effects on the differentiation of BMSCs, thus
   promoting the healing of irradiated mandibular bone defects. Taking
   together, these findings indicate that DCs may serve as a potential
   therapeutic target for radiation-induced jaw injury.

Materials and methods

   The work has been reported in line with the ARRIVE guidelines 2.0.

Animal

   Wild-type Sprague Dawley (SD) and F344 rats were purchased from Jihui
   Company (Shanghai, China). Male rats aged 4 weeks were used for the in
   vitro experiments. Male rats aged 10 weeks were used for the animal
   model and animals were randomly assigned to all experimental groups.
   All the rats were housed under specific pathogen-free (SPF) conditions
   with a circadian rhythm of 12 h. All the experiments received ethical
   approval from the Animal Care and Experiment Committee of Ninth
   People’s Hospital Affiliated to Shanghai Jiao Tong University School of
   Medicine (SH9H-2022-A799-SB).

Rat model

   A medical linear accelerator (Elekta Synergy) was applied for 6MVp
   X-ray radiation delivery. The operation was taken 100 cm away from the
   source with an incident beam on the mandibles of rats.

   For the radiation-induced jaw injury model, the rats were randomly
   assigned to two groups: control group (n = 10) and irradiated group
   (n = 10). For the irradiated group, SD rats were generally anesthetized
   with tribromoethanol (400 mg/kg, i.p.). A single dose of 20 Gy
   (6 Gy/min) IR was delivered to the left mandibles. One week after RT,
   the first molar of the left mandible was extracted. The control group
   received sham IR and also underwent the dental procedure. The rats were
   sacrificed 3 and 5 weeks after jaw RT for mandibles and cervical lymph
   nodes (LNs).

   To construct the irradiated mandibular defect rat model, a single dose
   of 10 Gy IR was delivered to the right mandibles of F344 rats. The
   control rats (Ctrl group) were also anesthetized and received sham IR.
   One week after RT, a 1 cm skin incision was made at the lower edge of
   the mandibles of all rats, and the masseter muscles were meticulously
   dissected, followed by an incision of the periosteum. A full-thickness
   penetrating critical-sized defect with 4 mm diameter was made by a
   trephine drill. Subsequent to the cleansing of the surgical site with
   saline, the mandible defects were loaded with 20 µL gelatin
   methacryloyl (Gelma, EFL-GM, China) containing 1 × 10^6 mDCs (IR + mDC
   group), tolDCs (IR + tolDC group) from F344 rats or without cells (IR
   group) (n = 5 per group). The surgical site was then sealed utilizing a
   suture. Unirradiated mandibles were loaded with blank Gelma at the same
   time. Nine weeks after RT, the mandibles were collected and fixed with
   4% formaldehyde for subsequent analyses. The major organs, including
   the livers, kidneys, lungs, hearts, and spleens, were also dissected
   for biosafety validation.

   The rats were sacrificed by inhaling CO[2]. We placed the rats in a
   pre-filled CO₂ chamber (20% displacement rate) and maintained exposure
   for 5 min after respiratory arrest. Then we confirmed death via
   bilateral thoracotomy.

Micro-computed tomography (micro-CT)

   Rat mandible samples were placed into a sample holder and scanned using
   a micro-CT instrument (PINGSENG Healthcare, China) at a resolution of
   18.26 μm at 40 kV and 250 µA. The morphology of bones was measured
   using Avatar 3 software. Bone parameters, including bone volume
   fraction (BV/TV), bone mineral density (BMD), trabecular thickness
   (Tb.Th), and trabecular number (Tb.N) were quantified.

Histology

   The bone tissues were decalcified in 10% EDTA at room temperature for 2
   weeks. All specimens were paraffin-embedded and then sectioned at a
   thickness of 5 μm for further hematoxylin-eosin (H&E), masson,
   immunohistochemistry (IHC), and immunofluorescence (IF) staining.

   H&E staining was performed to visualize the morphology and regeneration
   status of the tissues. Masson staining was performed to indicate
   collagen deposition and osteoid formation.

   For IHC and IF staining, sections were performed heat-mediated antigen
   retrieval in sodium citrate buffer for 20 min. The samples were then
   blocked in 5% bovine serum albumin solution for 1 h and incubated
   overnight at 4 °C with primary antibodies including CD11c (MBS25560,
   Mybiosource, USA), CD86 (sc-28347, Santa Cruz, USA), runt-related
   transcription factor 2 (RUNX2; [57]GB115631, Servicebio, China), CD4
   (ab237722, Abcam, USA), OCN ([58]GB115684, Servicebio, China).
   Subsequently, IHC sections were incubated with a secondary antibody
   conjugated to horseradish peroxidase (HRP) for 1 h and visualized by
   3,3’-diaminobenzidine (DAB) substrates. IF samples were incubated with
   corresponding secondary antibodies, including Alexa Fluor 488- and
   Cy3-labeled secondary IgG (Servicebio, China) for 1 h at room
   temperature. DAPI was used to counterstain the nuclei.

   The sections were imaged via a Zeiss Scope. A1 microscope (Carl Zeiss,
   Germany).

Quantitative real-time polymerase chain reaction (qRT-PCR)

   Total mRNA was extracted from the cells and mandible tissues using
   TRIzol reagent (Takara Biotechnology, Japan) and then
   reverse-transcribed to cDNA from 1000 ng of total RNA using Hieff
   Unicon^® V Universal Multiplex One Step qRT‒PCR Probe Kit (Yeasen
   Biotechnology, China). The qPCR system was performed on a Light
   Cycler-^® 480 Instrument II (Roche, Switzerland) using SYBR Green
   Master Mix (Yeasen Biotechnology, China). Primers used in this study
   are listed in Table [59]S1. Glyceraldehyde-3-phosphate dehydrogenase
   (Gapdh) was used as the housekeeping gene.

Isolation and IR of bone marrow-derived dendritic cells (BMDCs)

   BMDCs were isolated from the bone marrow of 4-week-old male SD and F344
   rats euthanized by inhaling CO[2]. Bone marrow single-cell suspensions
   were generated from tibias and femurs. After lysing erythrocytes, cells
   were resuspended in RPMI-1640 containing 10% fetal bovine serum (FBS)
   (Cyagen, USA), 100 U/ml penicillin/streptomycin, 20 ng/mL granulocyte
   macrophage-colony stimulating factor (GM-CSF), and 10 ng/mL
   interleukin-4 (IL-4) (PeproTech, USA) at a density of 5 × 10^5 cells/
   mL at 37 °C under 5% humidified CO[2] for 7 days. The medium was
   changed on day 3 and 6 with fresh growth factors. On day 7, floating
   and loosely adherent cells were harvested as immature DCs. BMDCs were
   also exposed to various IR doses (0, 2, 4, 8, 12, 16 Gy at 6.0 Gy/min)
   from 6MVp X-ray and cultured further for another 24 h. To obtain mDCs,
   100 ng/mL lipopolysaccharides (LPS) (Sigma-Aldrich, USA) were added to
   the medium concomitantly with IR.

   To induce tolDCs, 1,25(OH)[2]D[3] (10 nM; Sigma-Aldrich, USA) dissolved
   in anhydrous ethanol was added to the culture medium throughout the
   whole period of DC induction and then stimulated by a single 8 Gy dose
   of IR and LPS. The control cells were induced with an equal volume of
   anhydrous ethanol.

Isolation of BMSCs

   4-week-old SD rats were used for BMSCs isolation. The femur and tibia
   were removed aseptically after the rats were euthanized by inhaling
   CO[2]. Bone marrow was flushed out and cultured in Dulbecco’s modified
   Eagle’s medium (DMEM)/low glucose with 10% FBS and 100 U/mL
   penicillin/streptomycin. The attached cells were passaged when cell
   density reached 90%. Passage 2–3 BMSCs were used for cell experiments.

Preparation and collection of conditioned medium (CM) of BMDCs

   After IR and LPS stimulation for 24 h, mDCs were thoroughly washed with
   phosphate buffer saline (PBS) and cultured for another 24 h with fresh
   RPMI-1640 containing 10% FBS. The culture medium was centrifuged at 300
   xg for 10 min, followed by supernatant collection. Supernatants were
   mixed with DMEM at a ratio of 1:1 as BMDC-conditioned medium (CM) to
   culture BMSCs in vitro. BMSCs cultured in DMEM complete medium were
   used as controls.

Cell proliferation

   After exposure to IR for 24 h, imDC and mDC were incubated with 10 µL
   of cell counting kit-8 (CCK-8; Dojindo Molecular Technologies, Japan)
   and 90 µL of RPMI-1640 culture medium at 37 ℃ for 1 h before measuring
   the absorbance at 450 nm to quantify the proliferation activity of
   BMDCs. In another experiment, a CCK-8 assay was performed to check the
   viability of BMSCs co-cultured with DC-CM on day 1, day 3, and day 5.

Flow cytometry analysis (FCA)

   The expression of surface markers in BMDCs was detected 24 h after IR
   by flow cytometry using antibodies as follows: Zombie Violet
   (BioLegend, USA), FITC anti-rat CD11c (Santa Cruz, USA), APC anti-rat
   major histocompatibility complex class II (MHCII; eBioscience, USA),
   APC anti-rat CD40 (eBioscience, USA), BV421 anti-rat CD80 (BD
   Pharmingen, USA) and PE anti-rat CD86 (BioLegend, USA). The cells were
   incubated with their respective antibodies for 40 min at 4 ℃ and
   subsequently analyzed on a BD FACSAria III flow cytometer (BD
   Biosciences, USA). Data analyses were performed by FlowJo software.

Apoptosis analysis

   Apoptosis was quantified using the Annexin-V FITC apoptosis detection
   kit (BD Pharmingen, USA) according to the manufacturer’s instructions.
   Briefly, BMDCs were collected and resuspended in binding buffer 24 h
   after IR. 5 µL of FITC-Annexin V and 5 µL of propidium iodide (PI) were
   added to the buffer for 15 min of staining. The cells were analyzed by
   flow cytometry within 1 h.

ELISA

   The concentration of IL-1β, IL-6 IL-10, tumor necrosis factor (TNF)-α,
   transforming growth factor (TGF)-β, IL-12, and interferon (IFN)-γ in
   supernatants of BMDCs were measured 24 h after exposure to IR by
   enzyme-linked immunosorbent assay (ELISA) kits (Multi Sciences, China)
   following the manufacturer’s instructions.

Western blotting (WB)

   One hour after IR and LPS treatment, BMDCs were collected and lysed in
   radio immunoprecipitation assay lysis buffer with
   phenylmethanesulfonylfluoride and phosphatase inhibitors. For
   electrophoresis, 10% sodium dodecyl sulfate‒polyacrylamide gel
   electrophoresis gels were added with an equal amount of protein and
   electrically transferred to polyvinylidene fluoride membranes (Merck,
   Germany). The membranes were blocked, washed, and then incubated
   overnight with primary antibodies. The primary antibodies used in the
   current study were purchased from Cell Signaling Technology (USA) and
   listed as follows: anti-GAPDH (2118), α-Tubulin (2144),
   anti-phospho-P65 (pP65) (Ser536) (3033), anti-P65 (8242),
   anti-phospho-inhibitor of kappa B kinase α/β (pIKKα/β) (Ser176/180)
   (2697), anti-IKKα (2682), anti-phospho-inhibitory subunit of NF kappa B
   α (pIKBα) (Ser32) (2859), anti-IKBα (4812), anti-phospho-Janus kinase 2
   (pJAK2) (3776), anti-JAK2 (3230), anti-phospho-signal transducer and
   activator of transcription 3 (pSTAT3) (9145), anti-STAT3 (12640). Then,
   the membranes were incubated with horseradish peroxidase
   (HRP)-conjugated secondary antibodies for 1 h at room temperature. The
   signals were detected by chemiluminescence and semi-quantified through
   ImageJ software. The NF-κB pathway inhibitor JSH-23 (Abmole, USA) was
   added to the culture medium of BMDCs two hours before IR at the
   concentration of 20 µM to inhibit P65 nuclear translocation.

   The expression of osteogenic and lipogenetic proteins in BMSCs
   co-cultured with DC-CM were analyzed by WB. Protein bands were exposed
   to anti-osteocalcin (OCN; 20277-1-AP), anti-RUNX2 (20700-1-AP),
   anti-peroxisome proliferative activated receptor gamma (PPARG;
   16643-1-AP; Proteintech, China), and anti-alkaline phosphatase (ALP;
   abs158623; Absin, China) antibodies. The following steps had been
   previously outlined.

Migration assay

   After IR and LPS stimulation for 24 h, BMDCs were seeded into the upper
   chambers of the transwell inserts with a 8 μm pore membrane at the
   density of 2 × 10^5 /well. The lower chambers were RPMI-1640 with 100
   ng/mL chemokine-CC motif-ligand 19 (CCL19; PeproTech, USA). After one
   day, the upper chamber membranes were stained with crystal violet
   staining solution, and the number of migrating cells was observed under
   an optical microscope (Nikon, Japan).

Mixed leukocyte reaction (MLR)

   The BMDCs of each group were set as stimulator cells and incubated with
   mitomycin C (25 µg/mL) for 30 min at 37 ℃ to remove the proliferative
   ability. Responder cells were isolated from the spleens of F344 rats by
   lymphocyte separation medium (TBD science, China). Stimulator cells and
   responder cells were mixed at the ratio of 1: 10 and cultured for 72 h
   in round-bottomed 96 plates. The CCK-8 reagent was added to each well
   to detect the proliferation of lymphocytes. The absorbance was measured
   at 450 nm.

Live/dead cell staining

   A calcein/PI viability/cytotoxicity assay kit (Beyotime, China) was
   applied to assess the toxicity of CM to BMSCs as well as GelMA to DCs.
   On the third day of co-culture, 1 µL of calcein and 1 µL of PI were
   added to each well for 15 min of staining. The cells were observed and
   imaged by a fluorescence microscope.

RNA-sequencing (RNA-seq) and bioinformatics analysis

   The BMSCs were cultured with CM for 3 days. Total RNA was extracted and
   sent to DIATRE biological technology company (China) for library
   preparation, RNA-seq, and data analysis. DESeq2 package was carried out
   to perform differential expression analysis, and differentially
   expressed genes (DEGs) were determined by|log[2]FoldChange|≥1.5, and a
   P value < 0.05. Gene ontology (GO) analysis and Kyoto encyclopedia of
   genes and genomes (KEGG) pathway enrichment analysis were utilized to
   detect the molecular and biological processes as well as signaling
   pathways.

Detection of reactive oxygen species (ROS)

   One day after CM treatment, BMSCs were washed twice with PBS and
   incubated with 2,7-dichlorodi-hydrofluorescein diacetate (DCFH-DA;
   Beyotime, China) for 30 min at 37 ℃, and then imaged by a fluorescence
   microscope. The mean fluorescence intensity (MFI), which indicated the
   level of ROS, was measured by ImageJ software.

Scratch test

   Seeding BMSCs in a 6-well culture plate, allowing them to grow to full
   confluence. A sterile pipette tip was used to create 3 straight
   scratches across the cells. The cells were then washed with PBS to
   remove debris. The culture medium was replaced by low-serum (2%) CM.
   BMSCs were cultured in a 37 °C incubator and photographed by microscope
   at 0 and 24 h at the same position. The wound areas and migration rates
   in all groups were analyzed using ImageJ software.

ALP and Alizarin red S (ARS) staining

   BMSCs were co-cultured with mDC-CM and OriCell SD rat bone marrow
   mesenchymal stem cell osteogenic differentiation basal medium (Cyagen,
   USA). For ARS staining, the duration of osteogenic induction was 14
   days, and 7 days for ALP staining. The cells were washed twice using
   PBS, followed by fixation and staining with ARS kit (Solarbio, China)
   or BCIP/NBT alkaline phosphatase color development kit (Beyotime,
   China) for 30 min.

Oil red O staining

   BMSCs were co-cultured with DC-CM and OriCell SD rat bone marrow
   mesenchymal stem cell adipogenic differentiation basal medium (Cyagen,
   USA) for 7 days. Oil red O staining solution (Cyagen, USA) was utilized
   for staining following the manufacture’s protocols, and images were
   captured by an optical microscope.

Cell seeding in hydrogels

   First, mDCs and tolDCs were mixed with the GelMA solution at the
   concentration of 1 × 10^6 cells/mL, then light cured using ultraviolet
   light (405 nm) for 10 s and cultured by RPMI-1640 complete medium
   containing 10% FBS for 5 days.

Characterization of DCs loaded GelMA

   The GelMA was lyophilized using FreeZone (Labconco, USA), and the
   internal morphology was observed by scanning electron microscopy
   (GeminiSEM 300, ZEISS, Germany). The rheological analysis of GelMA or
   mDCs/tolDCs encapsulated GelMA was performed by a DHR-2 rheometer (TA,
   USA). The modulus change was recorded at a fixed angular frequency of
   1 rad/s and shear strain of 1%, respectively.

Statistical analysis

   All the data were obtained from at least three independent experiments.
   The values were presented as mean ± standard deviation (SD). The
   Student’s t-test and one-way ANOVA were used to perform statistical
   analyses using GraphPad Prism 7.0 (GraphPad Software, USA). P < 0.05
   was considered statistically significant.

Results

The radiation-induced jaw injury rat model shows impaired osteogenesis and
DCs activation

   To investigate the subacute and chronic jawbone complications caused by
   IR, we constructed a rat radiation-induced jaw injury model by a single
   dose of 20 Gy IR combined with mandibular tooth extraction
   (Fig. [60]1A). Rats were closely monitored over 5 weeks. The irradiated
   rats suffered weight loss, shedding, and impaired gingival healing
   (Supplementary Fig. [61]S1A-C). As seen in the sagittal and coronal
   plane of micro-CT, the bone density of tooth extraction sockets in the
   control groups increased over time. While there was almost no new bone
   formation at week 3 and a small amount of osteogenesis at week 5 in
   irradiated rats. Quantitative analysis of micro-CT results corroborates
   these observations. The values of BV/TV, BMD, and Tb.Th in the control
   group gradually increased over time. A statistically significant
   difference was observed between the CT values of radiation and
   unirradiated groups, and there was no difference between the irradiated
   groups at two time points (Fig. [62]1B-C). H&E staining indicated the
   emergence of nascent bones across control groups. The tooth socket was
   replaced by fibrous tissue in irradiated mandibles. Moreover, we
   observed sequestra with empty osteocyte lacunae and adipocytes in bone
   marrow at week 5, which was known as bone marrow steatosis
   (Fig. [63]1D). The changing trends of osteogenesis-related and
   lipogenesis-related genes, including Alp, osteopontin (Opn), Ocn,
   Runx2, Osterix, collagen I (Col1) and Pparg, were consistent with
   micro-CT and histological changes (Supplementary Fig. [64]S1D). RUNX2
   was a key transcription factor regulating early orthogenic
   differentiation. Its expression in the irradiated mandibles was
   significantly lower than that in the control group and gradually
   decreased over time (Supplementary Fig. [65]S1E-F). All these results
   suggest that bone regeneration was suppressed, and bone marrow
   structure was altered in the radiation-induced jaw injury model.

Fig. 1.

   [66]Fig. 1
   [67]Open in a new tab

   Phenotypic transformation of DCs in radiation-induced jaw injury rat
   model. (A) Flowchart of construction of radiation-induced jaw injury
   rat model. (B, C) Micro-CT and analysis of ROI (tooth extraction
   socket) in radiation-induced jaw injury rat model at week 3 and week 5.
   (D) Representative H&E staining images of tooth extraction socket and
   bone marrow area in the sagittal plane of the rat mandibles.
   (Newly-formed bone: NB; blue arrows: necrotic bone; yellow arrows:
   adipocytes) (E, F) Representative IF images showing the presence of
   CD11c^+CD86^+ DCs (white arrows) in the mandibles from rats with
   radiation-induced jaw injury, and the quantitative analysis of DC
   numbers. (G, H) Representative IHC images and quantification showing
   the presence of CD11c^+ DCs in the cervical lymph nodes from rats with
   radiation-induced jaw injury. (I, J) Representative IHC images and
   quantification showing the presence of CD4^+ T cells in the mandibles
   from rats with radiation-induced jaw injury. Data were expressed as
   means ± SD. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. Graphs
   were created with BioRender.com

   DCs can up-regulate MHC-II and co-stimulatory molecules after
   activation and maturation. To demonstrate the effect of IR on DC
   activity, we utilized multiplex immunofluorescence staining to localize
   CD11c^+CD86^+ cells as activated DCs. The staining and quantitative
   results showed that the number of double-positive cells in the bone
   marrow of irradiated mandible was significantly higher than that of the
   corresponding controls at both points (Fig. [68]1E-F). After
   recognizing and capturing antigens in peripheral tissues, DCs can
   migrate to the draining LNs, where they interact with naïve T cells,
   effectively induce adaptive immune responses, and further exacerbate
   inflammation. We performed IHC staining of the cervical LNs, the
   CD11c-positive area fraction of irradiated rats reached 8.2% and 15.5%
   at week 3 and 5 respectively, which was considerably higher compared
   with rats that received sham IR. The positive cells in the control
   group were mostly located in the lymphoid follicles. While in the
   irradiated rats, positive cells were located in both the lymphoid
   follicles and paracortex zones (T-cell area) (Fig. [69]1G-H).
   Additionally, IHC staining was performed on jawbones to label CD4^+ T
   cells. At week 3, there was no difference between the proportion of
   CD4^+ cells in the bone marrow between the experimental and control
   groups, which might be attributed to post-extraction inflammatory
   responses. By week 5, the control group exhibited a decline in the
   proportion of positive areas, whereas the irradiated group maintained
   high expression levels (Fig. [70]1I-J).

   To sum up, IR caused macroscopic and microenvironmental remodeling of
   the jawbone. Dendritic cells were activated, which might contribute to
   the amplification of inflammation and the progression of lesions.

DCs transform into a mature phenotype after IR in vitro

   To further validate the effects of IR on DCs in vitro, we first imposed
   a gradient dose of IR on BMDCs (Fig. [71]2A). One day after IR, a dose
   above 4 Gy reduced the proliferation rate of BMDCs dose-dependently and
   increased the proportion of apoptosis in imDCs and mDCs. However, there
   was no significant difference in apoptosis between 4 Gy and 8 Gy
   (Supplementary Fig. [72]S2A-C). Since patients with head and neck
   tumors often undergo up to 70 Gy IR, we would adopt the highest dose as
   possible. To ensure 50% cell viability, a single dose of 8 Gy was used
   for subsequent experiments.

Fig. 2.

   [73]Fig. 2
   [74]Open in a new tab

   Radiation-induced BMDC activation and maturation in vitro. (A)
   Flowchart of BMDC culture and irradiation process. BMDCs were
   stimulated with 100 ng/mL LPS and exposed to gradient irradiation. (B,
   C) Flow cytometry plots showing gating strategy for MHCII^+, CD40^+,
   CD80^+ and CD86^+ populations in CD11c^+ DCs, and their corresponding
   bar charts. (D) Bar charts illustrating the concentration of TNF-α,
   IL-6, IL-10, TGF-β, IL-1β, IL-12p70 and IFN-γ in supernatant of
   irradiated and LPS treated BMDCs by ELISA. (E) Representative western
   blotting images revealed the expression of NF-κB pathway-associated
   factors in irradiated and LPS treated BMDCs. (F) Relative protein
   expression of p-P65/P65, p-IKKα/β/IKKα and p-IKBα/IKBα. (G)
   Representative western blotting images of irradiated BMDCs treated with
   NF-κB pathway inhibitor JSH-23 from whole cell lysate. (H) Relative
   protein expression of p-P65/P65. (I) Flow cytometry analysis of
   CD11c^+MHCII^+ and CD11c^+CD86^+ cells in irradiated BMDCs treated with
   NF-κB pathway inhibitor JSH-23. Data were expressed as means ± SD.
   *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. Graphs were
   created with BioRender.com. Full-length blots of E, G are presented in
   Supplementary Fig. [75]S7

   One day after RT, DCs were detected for co-stimulatory molecules by
   flow cytometry. As indicated in the gating strategy, IR up-regulated
   the expression of CD40, CD80, CD86, and MHCII in both imDCs and mDCs
   (Fig. [76]2B-C). Pro- and anti-inflammatory cytokines in the
   supernatants of DCs were measured by ELISA. The imDCs secreted only a
   small amount of cytokines, and there was no difference with or without
   IR, except for TGF-β. For mDCs, more cytokines were secreted. IR
   significantly increased the concentration of TNF-α, and reduced the
   concentration of IL-6, IL-10, and TGF-β. However, IR didn’t alter the
   secretion level of IL-1β, IL-12p70, and IFN-γ. The changes in cytokine
   mRNA levels were consistent with ELISA results (Fig. [77]2D, S2D). In
   addition, IR also enhanced the migration of imDCs and mDCs in the
   presence of CCL19 (100 ng/mL) (Supplementary Fig. [78]S2E-F), as well
   as the ability of mDCs to stimulate lymphocyte proliferation
   (Supplementary Fig. [79]S2G), which were also the characteristics of DC
   activation.

   Several studies have revealed activation of the NF-κB and JAK-STAT
   signaling pathways in DCs across multiple infectious, allergic, and
   autoimmune conditions [[80]25–[81]27]. However, whether these two
   pathways involved in the radiation-induced DC maturation have not been
   fully elucidated. WB along with inhibitory experiments revealed the
   mechanism underlying the maturation of BMDCs induced by IR at the
   protein level. The phosphorylation levels of P65, IKKα/β, and
   inhibitory subunit of IkBa in imDCs and mDCs were significantly
   increased after RT, with imDCs lower than mDCs (Fig. [82]2E-F). The
   NF-κB pathway inhibitor JSH-23 restrained P65 nuclear translocation of
   irradiated imDC (Fig. [83]2G-H). Flow cytometry also showed that JSH-23
   decreased the MFI of CD86 and MHC-II (Fig. [84]2I). However, the
   JAK-STAT pathway failed to be activated under IR conditions. There was
   no significant difference of p-JAK2 and p-STAT3 expression compared to
   their total proteins observed in imDCs or mDCs (Supplementary Fig.
   [85]S3A-D).

   In brief, when IR was applied to imDCs solely, the cells were in a
   semi-mature state characterized by low levels of co-stimulatory
   molecules, migration, MLR capability, and cytokine production. BMDCs
   achieved full maturation when induced by both LPS and IR to aggravate
   local inflammation. To better elucidate the role of DC inflammatory
   response in radiation-induced jaw injury, mDCs were used for follow-up
   experiments.

Irradiated DC-CM induces dysfunction of BMSCs through inflammatory-reparative
imbalance and oxidative stress

   BMSCs are pluripotent stem cells with properties including
   multipotency, self-renewal, and immunoregulation, serving as the basis
   for bone regeneration. IR-activated DCs may reprogram the landscape of
   jawbone microenvironment, thus indirectly affecting BMSCs. To
   investigate the role of molecular signals from mDC on BMSCs, we
   cultured BMSCs with CM from irradiated/unirradiated mDCs. BMSCs
   cultured in CM-free medium were the controls (Fig. [86]3A).

Fig. 3.

   [87]Fig. 3
   [88]Open in a new tab

   Irradiated DC-CM impairs BMSC proliferation and migration through
   oxidative stress and inflammatory response. (A) Graphical
   representation of the experimental strategy for BMDCs culture medium
   (CM) and BMSCs co-culture system. BMDCs were stimulated with 100 ng/mL
   LPS and 8 Gy irradiation. The CM was collected the next day for BMSC
   culture. (B) Representative fluorescence images for live/dead staining
   of BMSCs cultured in CM from unirradiated/irradiated mDCs for 72 h. The
   green and red channels represent the Calcein-AM and PI staining of live
   and dead cells respectively. (C, D) GO and KEGG enrichment analysis of
   the differentially expressed mRNAs (|log[2]FoldChange|≥1.5, P
   value < 0.05, and P value adjusted using a false discovery rate)
   between BMSCs from mDC-8 Gy and mDC-0 Gy group on the third day of
   co-culture. (E) Effects of irradiated mDC-CM on proliferation of BMSCs
   detected by CCK8 on day 1, 3, 5. (F) qRT-PCR results of
   inflammation-related gene expression in BMSCs cultured in CM from
   unirradiated/irradiated mDCs for 72 h. (G, H) Reactive oxygen species
   detection and quantification of BMSCs cultured in
   unirradiated/irradiated mDC-CM. (I, J) Representative light microscope
   images showing the migration of BMSCs after scratches were applied to
   the cells cultured in unirradiated/irradiated mDC-CM for 24 h. The
   percentage of closed wound area was used to calculate the migration
   ability of each group. Data were expressed as means ± SD. *P < 0.05;
   **P < 0.01; ***P < 0.001; ****P < 0.0001. Graphs were created with
   BioRender.com

   Live/dead staining showed that BMSCs in both the control group and CM
   co-culture group remained alive and intact on day 3. Only a small
   amount of cell debris and dead cells were observed (Fig. [89]3B).

   RNA-seq was performed on BMSCs cultured in CM for 3 days. GO analysis
   revealed that the top 10 enriched terms for upregulated genes in BMSCs
   from mDC-8 Gy group compared mDC-0 Gy group included regulation of
   inflammatory response, positive regulation of response to external
   stimulus and wound healing, while GO terms enriched in downregulated
   genes included extracellular matrix and ossification (Fig. [90]3C).
   KEGG enriched in downregulated genes consisted of Wnt signaling pathway
   and ECM-receptor interaction (Fig. [91]3D). GO and KEGG analyses
   between the mDC-0 Gy and control groups were shown in Supplementary
   Fig. [92]S4. These results indicated a dysregulation between
   inflammatory and reparative responses in BMSCs. ROS production was
   significantly triggered by DC-CM treatment in an IR-dependent manner
   (Fig. [93]3G-H), accompanied by disordered paracrine signaling. BMSCs
   manifested increased mRNA expression of pro-inflammatory cytokines and
   chemokines, including Ccl6, Ccl20, Cxcl2, Cxcl3, Il1a, Il1b, and Il6
   (Fig. [94]3F).

   CCK-8 assay was performed on day 1, 3, and 5 after co-culture. DC-CM
   significantly decreased the proliferation of BMSCs. However, no
   difference was observed between the irradiated and unirradiated DC-CM
   treatment groups (Fig. [95]3E). Scratch assay suggested that CM
   co-culture could inhibit the migratory ability of BMSCs, irradiated
   DC-CM further reduces wound healing area (Fig. [96]3I-J).

Abnormal differentiation of BMSCs result from irradiated DC-CM

   Previous studies and our rat model have suggested that
   radiation-induced mandibular injury is characterized by suppressed
   osteogenesis and pathological adipocytic infiltration within the bone
   marrow [[97]28, [98]29]. Given the potential influence of immunological
   factors on cellular differentiation, we evaluated how mDC-CM regulates
   osteogenic and adipogenic differentiation balance of BMSCs.

   qPCR revealed that CM from irradiated mDCs suppressed the osteogenesis
   of BMSCs, reducing mRNA expression of Runx2, Alp, Opn, integrin binding
   sialoprotein (Ibsp) and Ocn of BMSCs (Fig. [99]4A). We found that CM
   decreased the ALP activity and mineralization of BMSCs compared to the
   control group by ALP and ARS staining, while irradiated DC-CM could
   aggravate the inhibition (Fig. [100]4B). WB supported the result of
   qPCR that irradiated DC-CM down-regulated osteogenesis-related proteins
   including RUNX2, ALP and OCN (Fig. [101]4C-D). Oil red O staining was
   adopted to assess the adipogenic differentiation of BMSCs
   (Fig. [102]4F-G). The application of post-radiation DC-CM increased
   adipogenic differentiation of BMSCs, the expression of the adipogenic
   marker genes like enhancer binding protein alpha (Cebpa), enhancer
   binding protein beta (Cebpb), perilipin (Plin1), and Pparg
   (Fig. [103]4E), as well as the lipogenesis-related protein PPARG
   (Fig. [104]4H-I). Taken together, irradiated DCs disturbed the
   adipogenic and osteogenic balance of BMSCs.

Fig. 4.

   [105]Fig. 4
   [106]Open in a new tab

   Irradiated DC-CM disrupts the osteogenic-adipogenic differentiation
   balance in BMSCs. (A) qRT-PCR results of osteogenesis-related gene
   expression in BMSCs treated with unirradiated/irradiated mDC-CM. (B)
   ARS staining (up) and ALP staining (down) of unirradiated/irradiated
   mDC-CM treated BMSCs. (C) Representative western blotting images of
   osteogenesis-related proteins in BMSCs cultured with
   unirradiated/irradiated mDC-CM. (D) Relative protein expression of
   ALPL, RUNX2, and OCN. (E) qRT-PCR results of lipogenesis-related gene
   expression in BMSCs treated with unirradiated/irradiated mDC-CM. (F, G)
   Oil O staining and quantification of unirradiated/irradiated mDC-CM
   treated BMSCs. (H) Representative western blotting images of
   lipogenesis-related protein in BMSCs cultured with
   unirradiated/irradiated mDC-CM. (I) Relative protein expression of
   PPARG. Data were expressed as means ± SD. *P < 0.05; **P < 0.01;
   ***P < 0.001; ****P < 0.0001. Full-length blots of C, H are presented
   in Supplementary Fig. [107]S8

TolDCs exhibit hyporesponsiveness to IR and reverse adverse effects of mDCs
on BMSCs

   We generated tolDCs through VitD3 induction, which is a well-validated
   protocol (Fig. [108]5A). TolDCs exhibited maturation resistance.
   qRT-PCR showed that irradiated tolDCs (tolDC-8 Gy) down-regulated gene
   levels of inflammatory cytokines Il1b and Tnf-α, and up-regulated
   anti-inflammatory factor Il-10 than unirradiated tolDCs (tolDC-0 Gy)
   (Fig. [109]5B). After IR, the percentage of MHC-II^+ DCs did not change
   in tolDCs as detected by flow cytometry. Although IR increased the
   expression of CD40, CD80, and CD86 in tolDCs, the magnitude of this
   activation was significantly lower than that of the irradiated mDCs
   (Fig. [110]5C). The phosphorylation levels of P65, IKKα/β, and IkBa in
   irradiated tolDCs were lower than mDCs. There was no statistical
   difference between unirradiated cells (Fig. [111]5D-E).

Fig. 5.

   [112]Fig. 5
   [113]Open in a new tab

   TolDCs exhibit radioresistance and have capacity to counteract the
   effect of mDCs on BMSCs. (A) Flowchart of inducing tolDCs with VitD3.
   VitD3 was added to the culture medium throughout the whole period of DC
   induction. tolDCs were then stimulated with 100 ng/mL LPS and exposed
   to 8 Gy irradiation. (B) qRT-PCR results of inflammation-related gene
   expression in unirradiated/irradiated tolDCs. (C) Flow cytometry
   analysis of MHCII^+, CD40^+, CD80^+ and CD86^+ populations in
   unirradiated/irradiated mDCs and tolDCs. (D, E) Representative western
   blotting images of unirradiated/irradiated mDCs and tolDCs. Relative
   protein expression of p-P65/P65, p-IKKα/β/IKKα and p-IKBα/IKBα. (F) ARS
   staining (up) and ALP staining (down) of BMSCs cultured with
   unirradiated/irradiated mDC-CM and tolDC-CM. (G) Representative western
   blotting images of osteogenesis-related proteins in BMSCs cultured with
   unirradiated/irradiated mDC-CM and tolDC-CM. (H) Relative protein
   expression of ALPL, RUNX2 and OCN. (I, J) Oil O staining and
   quantification of BMSCs cultured with unirradiated/irradiated mDC-CM
   and tolDC-CM. (K) Representative western blotting images of
   lipogenesis-related protein in BMSCs cultured with
   unirradiated/irradiated mDC-CM and tolDC-CM. (L) Relative protein
   expression of PPARG. Data were expressed as means ± SD. *P < 0.05;
   **P < 0.01; ***P < 0.001; ****P < 0.0001. Full-length blots of E, G, K
   are presented in Supplementary Fig. [114]S9

   We also collected tolDC-derived CM (tolDC-CM) to induce the
   differentiation of BMSCs. TolDCs restored the osteogenic
   differentiation ability of BMSCs. ALP and ARS staining results showed
   that irradiated tolDCs induced more ALP activity and mineralization
   than mDCs (Fig. [115]5F). WB revealed upregulated expression of
   osteogenic marker proteins (ALPL, RUNX2, and OCN) in the tolDC-8 Gy
   group compared to the mDC-8 Gy group. However, RUNX2 and OCN levels in
   the tolDC-8 Gy group were still lower than those in the mDC-0 Gy group
   (Fig. [116]5G-H). We also observed less lipid droplet accumulation in
   tolDC-CM treated BMSCs (Fig. [117]5I-J). The protein level of PPARG was
   down-regulated as well (Fig. [118]5K-L).

TolDCs accelerate bone healing in irradiated mandibular defect rat model

   To validate that excessive inflammation of DCs aggravate
   radiation-induced jaw injury, and that the inhibition of DC maturation
   may repair or even reverse the lesion in vivo, an irradiated mandibular
   defect rat model was constructed. GelMA loaded with tolDCs/mDCs was
   placed into the defect area (Fig. [119]6A). As demonstrated in
   Supplementary Fig. [120]S5A, the GelMA exhibited a porous network
   structure. The rheological analysis confirmed that the storage modulus
   (Gʹ) of GelMA was significantly higher than loss modulus (Gʹʹ). The
   modulus of GelMA, as a function of time, was not significantly impacted
   by the DC encapsulation (Supplementary Fig. [121]S5B). Both tolDCs and
   mDCs encapsulated within GelMA maintained viability (Supplementary Fig.
   [122]S5C).

Fig. 6.

   [123]Fig. 6
   [124]Open in a new tab

   Local transplantation of tolDCs promotes bone healing of irradiated
   mandibular defects in rats. (A) Flowchart of irradiated mandibular
   defect rat model construction and mDC/tolDC therapy. (B, C) Micro-CT
   coronal images (top) and 3D reconstruction images (bottom) of
   irradiated rat mandibular defect repair after local transplantation of
   mDCs and tolDCs. (C) Corresponding measurements of BMD and BV/TV show
   the healing efficiency of mandible defects in rats. (D) Representative
   H&E staining of mandible defects loaded with mDCs and tolDCs, showing
   the histological morphology of newly-formed bone (NB; black), bone
   marrow (MB; red), vessel (V; green), sequestrum (S; orange) and GelMA
   (blue). (E) Representative Masson staining of the mandible defects
   loaded with mDCs and tolDCs, showing the histological morphology of
   newly-formed bone (NB) and mature bone (MB). (F, G) Representative IHC
   images and quantification of showing the presence of OCN^+ osteoblasts
   in the mandible defects loaded with mDCs and tolDCs. Data were
   expressed as means ± SD. *P < 0.05; **P < 0.01; ***P < 0.001;
   ****P < 0.0001. Graphs were created with BioRender.com

   According to the radiographs, 9 weeks after RT, the IR group displayed
   less bone deposition as compared to the control group. The IR + tolDC
   group possessed stronger periosteal reaction and bone regeneration
   ability than IR group, while there was nearly no new bone formation in
   IR + mDC group. The quantitative analysis of BV/TV revealed the highest
   values in the control group, which were 1.69 and 3,73 times higher than
   that of the IR and IR + mDC groups respectively. There was no
   statistical difference in BV/TV values between the control group and
   IR + tolDC group. Additionally, tolDC-encapsulated hydrogel
   significantly increased the BMD of the newly-formed mandible than the
   IR and IR + mDC group, achieving levels comparable to the control group
   (Fig. [125]6B-C).

   HE staining of major organs showed that the irradiated rats suffered
   from renal cortical and glomerular atrophy. There were no obvious
   morphological changes of organs among irradiated rats, indicating the
   satisfactory safety profile of the cell therapy (Supplementary Fig.
   [126]S6). HE staining also showed substantial newly-regenerated bone
   extending from the defect margin toward the central region in both the
   control and IR + tolDC groups. The new bones had numerous bone marrow
   and vascular-like structures, suggesting robust osteogenic activity. In
   contrast, the IR group displayed limited bone regeneration accompanied
   by fibroblast-like components at the margin of the defect. Unabsorbed
   GelMA was encapsulated within fibrous tissue in this group. As for the
   IR + mDC group, there was an absence of new bone formation, and there
   were no osteocytes observed in bone lacuna in some bone fragments,
   namely the sequestrum (Fig. [127]6D). Masson staining indicated the
   emergence of nascent bone in the IR and IR + tolDC groups, while
   control group exhibited more depositions of mature bone within the
   defect area (Fig. [128]6E). OCN was an osteogenic marker expressed in
   mature bone cells and localized in the newly formed bone tissue. A
   significant reduction in OCN-positive areas was observed in both the IR
   and IR + tolDC group compared to the control cohort. Although tolDC
   administration partially rescued OCN expression compared to IR
   counterparts, the restored levels remained suboptimal relative to the
   control ones (Fig. [129]6F-G).

Discussion

   Growing evidences have confirmed the importance of the immune
   microenvironment in bone repair [[130]30–[131]32]. High doses of IR
   promotes the recruitment and activation of different immune cells, thus
   breaking the homeostasis of the osteoimmune microenvironment, resulting
   in radiation-induced jaw injury and eventually osteoradionecrosis
   [[132]12, [133]33].

   When exposed to diverse pathogenic stimuli, DCs exhibit either
   activated or suppressed states, modulating both inflammatory responses
   and adaptive immune responses, which may either exacerbate or relieve
   inflammatory jawbone diseases [[134]34, [135]35]. In this study, we
   intended to elucidate the role of dendritic cells in disease
   progression through the utilization of a well-established rat ORNJ
   model and cell experiments.

   A notable elevation of mature DC percentage within the irradiated
   jawbone marrow was observed compared to controls. Simultaneously, the
   irradiated rats exhibited more DCs in cervical lymph nodes concurrent
   with CD4⁺ T cell infiltration in the jawbone, implying DC involvement
   and functional activation, which might explain the disruption of immune
   homeostasis in the jawbone.

   BMDCs cultured with GM-CSF and IL-4 were used to investigate the effect
   of IR on DCs in vitro. BMDCs exhibit cellular heterogeneity [[136]36].
   Some studies suggested that BMDCs resembled monocyte-derived
   inflammatory dendritic cells [[137]37]. Helft et al. [[138]38]
   performed unsupervised hierarchical clustering of the transcriptome of
   GM-CSF-induced DCs (GM-DC), revealing that GM-DC clustered most closely
   with migratory DCs, characterized by specific expression of
   maturation-related, migratory, and immunomodulatory transcripts
   [[139]39, [140]40]. Therefore, BMDCs only represent part of the DC
   population in vivo.

   IR-induced BMDCs acquired a unique semi-mature state, characterized by
   increased surface expression of the MHC-II and co-stimulatory
   molecules, coupled with enhanced migratory capacity through the NF-κB
   axis, while maintaining baseline levels of cytokine secretion and
   limited potential in MLR assay. Bacteria-derived LPS provided critical
   secondary signals for the complete maturation of DCs, accompanied by
   increased secretion of pro-inflammatory cytokines. Meyer proposed the
   theory of ‘radiation, trauma, and infection’, suggesting that bacterial
   infection is a primary factor of ORNJ [[141]9]. Post-RT dental
   extraction provides a favorable environment for bacterial colonization
   by oral pathogens such as Actinomyces spp., Streptococcus intermedius,
   etc [[142]41–[143]43], which greatly increases the incidence of
   osteomyelitis and osteonecrosis [[144]44]. DCs, as the sentinels of the
   immune system, can sense and alert the body to pathogens or danger
   signals [[145]45]. In radiation-induced jaw injury complicated by
   secondary infection, hyperactivation of DCs is likely to exacerbate
   regional inflammation, establishing a self-perpetuating circle of
   inflammation-necrosis-infection [[146]46].

   Radiation-induced immune dysregulation seriously disrupts bone
   homeostasis, particularly affecting the function of BMSCs. For example,
   the expansion of B cells and CD8⁺ T cell subsets after RT impaired BMSC
   osteogenesis, leading to systemic bone loss [[147]47]. DCs serve as
   indispensable regulators within the osteoimmune microenvironment,
   establishing bidirectional modulatory control over bone remodeling
   through intercellular crosstalk with cells of the skeletal system
   [[148]15, [149]21, [150]48]. DC depletion directly hindered the
   heterotopic ossification of biomaterial [[151]49]. In our study, we
   found that irradiated mature DCs and DC-derived secretive cell factors
   triggered inflammatory responses in BMSCs, thereby suppressing
   proliferation, migratory ability, and osteogenic differentiation while
   enhancing adipogenic differentiation. These alterations were in
   accordance with the compositional remodeling observed in the bone
   marrow of radiation-induced jaw injury rat models. Although IR can
   impose direct cytotoxic effects on stromal cells, DCs further amplify
   the adverse effects on BMSCs. However, in the current study, we only
   discussed the synergistic effects of multiple cytokines of mDCs. The
   mechanistic contributions of a specific cytokine remain to be fully
   elucidated.

   We hypothesized that by suppressing inflammatory activation of DCs,
   radiation-induced injury could be ameliorated. Adoptive cell transfer
   using patients’ own tolDCs has been successfully translated to the
   clinic to treat autoimmune or chronic inflammatory diseases [[152]50,
   [153]51]. It has been reported that tolDCs can be generated using
   additional agents or lentiviral vector-based approaches
   [[154]52–[155]55]. Lentiviral transduction remains technically
   challenging. Previous studies have demonstrated modest efficiency
   (30–40%), or required higher vector doses—a strategy that might trigger
   undesired DC maturation or cell toxicity [[156]56, [157]57]. VitD3 is a
   potent natural modulator of both innate and adaptive immunity,
   selectively suppressing the NF-κB transcription in cDCs with adequate
   safety and efficacy to develop clinical-grade tolerogenic tolDCs
   [[158]58, [159]59]. GelMA has been widely used in tissue engineering to
   provide a physiologic microenvironment for transplanted cells because
   of its compositional similarity to natural extracellular matrices
   [[160]60, [161]61]. We verified that the tolDCs encapsulated GelMA
   effectively promoted bone regeneration in irradiated jaw defects.
   Regarding the therapeutic mechanisms of tolDCs according to recent
   studies, they promote hyporesponsiveness of T cells [[162]62, [163]63].
   Besides, tolDCs impart a robust anti-inflammatory stimulus that
   mitigates inflammation through a by-stander effect [[164]64].
   Nevertheless, the underlying molecular mechanisms of tolDCs therapy
   remain to be fully elucidated. Future studies should also test
   different cell doses on hydrogel to optimize therapeutic effect.

Conclusion

   To summarize, our study highlights that the hyperactivation and
   maturation of DCs served as a potential pathogenic factor in
   radiation-induced jaw injury. DCs exacerbated the local inflammatory
   responses, induced oxidative stress, and abrogated the biological
   functions of BMSCs. The administration of tolDCs greatly promoted the
   progression of jawbone regeneration post-IR, suggesting that tolDCs
   might be a promising therapeutic strategy against radiation-induced jaw
   injury (Fig. [165]7).

Fig. 7.

   [166]Fig. 7
   [167]Open in a new tab

   Schematic conceptualization of this study. IR activated the NF-κB
   signaling pathway to promote activation and maturation of DCs,
   amplifying inflammatory signals through the release of pro-inflammatory
   cytokines. The mDCs triggered oxidative stress of BMSCs, showing
   increased intracellular ROS levels, inhibited osteogenic
   differentiation, and enhanced adipogenic differentiation, which
   accelerated the progression of radiation-induced jaw injury.
   VitD3-induced tolDCs exhibited IR resistance and anti-inflammatory
   properties, possessing the therapeutic potential to accelerate bone
   regeneration in irradiated jaw defects through local administration.
   (IR: ionizing radiation; mDC: mature dendritic cell; tolDC: tolerogenic
   dendritic cell; BMSC: bone marrow mesenchymal stem cell; ROS: reactive
   oxygen species; vitamin D3: vitD3)

Electronic supplementary material

   Below is the link to the electronic supplementary material.
   [168]Supplementary Material 1^ (4.1MB, docx)

Acknowledgements