Male C57BL/6J (H-2D^b) and BALB/c (H-2D^d) mice, 6–8 weeks of age and weighing 20–25 g, were purchased from The Jackson Laboratory (Clea Japan, Tokyo, Japan). The mice were kept for at least one week on a 12-h light and dark cycle. All experiments were approved by the Animal Care and Use Committee of Kyushu University (Permit Number: A24-033-0) and were conducted in strict compliance with ethical guidelines.

The cloning and expression of a glutathione S-transferase full-length M180 amelogenin fusion construct and the purification of rM180 have been described previously (14). The full-length mouse amelogenin (M180) cDNA was inserted into a vector and transformed into competent Escherichia coli. Bacterial pellets containing recombinant glutathione S-transferase-rM180 were harvested, and the fusion protein was cleaved on a column using PreScission protease (GE Healthcare, Boston, MA, USA) to obtain purified rM180. The removal of endotoxins from rM180 was verified (endotoxin level: < 0.03 EU per 10 µg of rM180).

A 1.0 × 1.0 cm² piece of skin from the back of a C57BL/6J mouse was transplanted onto the back of a BALB/c mouse. The recipient BALB/c mice were randomly divided into two groups: rM180 and control mice. Transplanted skin was treated with either PBS or rM180. The purified rM180 was adjusted to a concentration of 10 μg/100 μL with PBS, and 100 μL of the solution was applied dropwise evenly to the surface of the graft to be attached to the recipient. The transplanted skin was sutured at four points around the periphery using ETHICON COATED VICRYL^® Plus Antibacterial (polyglactin 910) Suture (Ethicon, Somerville, NJ, USA) as soon as possible after administered. From day 7 onward, the allografts were evaluated daily for skin necrosis by at least two observers in a blinded manner. Allograft rejection (necrotic score 0) was defined as spontaneous graft detachment. The necrotic areas (black spots) were evaluated according to the previous study by Zhao et al. (15). Briefly, they were roughly estimated by visual inspection with ImageJ 1.53 (NIH) as a supplementary tool, and six different score levels were defined according to the percentage of the necrotic area of the graft.

To investigate the role of Lcn2, the mice received subcutaneous injections of either anti-mouse lipocalin 2/NGAL monoclonal antibody ([MAB1857]; R&D Systems, Minneapolis, MN, USA) or rat IgG2A isotype control antibody ([MAB006]; R&D Systems). Antibodies were diluted in sterile PBS, and 2 μg in a total volume of 4 μL was administered at four points around the graft site every 24 h for seven consecutive days, starting 24 h after transplantation.

Paraffin sections (thickness, 10 µm) of the skin graft tissues were deparaffinized using xylene and dehydrated with ethanol. Sections were stained with H&E or immunohistochemistry. Non-specific staining was blocked by incubation with Blocking One Histo (Nacalai Tesque, Kyoto, Japan) for 30 min at room temperature. These slides were incubated with the primary antibody, anti-rabbit CD4 antibody (ab287724; Abcam, Cambridge, UK), at 1:20 dilution, Anti-mouse CD8 alpha antibody (sc-7970, Santa Cruz Biotechnology, Dallas, TX, USA) at 1:250 dilution, Anti-MHC class II ([MRC OX-6]; Abcam) at 1:1000 dilution, CD19 Monoclonal antibody ([6OMP31]; Invitrogen, Carlsbad, CA, USA) at 1:500 dilution, F4/80 Monoclonal antibody ([BM8]; Thermo Fisher Scientific™,Waltham, MA, USA) at 1:500 dilution, Normal rat IgG (sc-2026, Santa Cruz Biotechnology) at 1:500 dilution, Mouse (G3A1) mAb IgG1 isotype control (5415S, Cell Signaling Technology) at 1:500 dilution, Rabbit (DA1E) mAb IgG XP isotype control (3900S, Cell Signaling Technology) at 1:500 dilution, Anti-Ym-1 + Ym-2 (Chil4) antibody ([EPR15263]; Abcam) at 1:200 dilution, Proteintech NGAL (Lcn2) polyclonal antibody (Proteintech Group Inc, Wuhan, China) at 1:400 dilution, Anti-OCT2 (Pou2f2) antibody (Sigma-Aldrich, St. Louis, CA, USA) at 1:500 dilution, or purified anti-Arginase1 antibody ([O94E6]; Biolegend, San Diego, CA, USA) at 1:200 dilution overnight at 4 °C in the dark. They were washed and incubated with a secondary antibody, and the nucleus was stained using SlowFade™ Diamond Antifade Mountant with DAPI (Thermo Fischer Scientific™). Photographs were taken using a BZ8000 (Keyence Co., Osaka, Japan), and the numbers of CD4, CD8, CD19, F4/80, MHCII, Chil4, Lcn2, and Pou2f2-positive cells were counted using a hybrid cell count application with BZ-X Analyzer software (Keyence Co). Images were analyzed using ZEISS LSM700 (Carl Zeiss, Oberkochen, Germany) and ZEN 2012 software. H&E staining was performed on allografts to assess tissue morphology.

Blood was collected from the facial vein using an Animal Lancet 5 mm (AS ONE, #21328703) and centrifuged at 2000 × g for 10 min to aspirate the serum. The spleens were passed through 70 μm cell strainers and centrifuged. The spleens were subjected to a round of red blood cell lysis. RBC Lysis Buffer (Biolegend, #420302) was used to lyse the erythrocytes. Skin grafts were harvested on day 4 or 6 post-transplant and processed into single-cell suspensions using Dri Tumor & Tissue Dissociation Reagent (BD Horizon™, # 661563) (BD Biosciences, San Diego, CA, USA).

Serum cytokine levels in peripheral blood were measured using the BD™ Cytometric Bead Array (CBA) Mouse Th1/Th2/Th17 CBA Kit purchased from BD Pharmingen™ (BD Biosciences).

Freshly isolated spleen cells were obtained by gently milling the mouse spleens in PBS. Single-cell suspensions were washed with PBS and stained with live/dead fixable viability stain (Thermo Fisher Scientific™, # L34961). Fc receptors were blocked using Fc block (Biolegend, #156604) before surface staining with antibodies of interest ( Supplementary Table 1 ) in FACS wash buffer (Biolegend, #420201; 10 min, 4 °C). The cells were washed and fixed with 1% formaldehyde. For intracellular cytokine or transcription factor assessment, cells were fixed and permeabilized with Cyto-Fast™ Fix/Perm Buffer Set (Biolegend, #426803) or Biolegend True-Nuclear™ Transcription Factor Buffer Set (Biolegend, #424401) before staining with antibodies targeting markers of interest. Isotype controls were used to confirm antibody specificity. The cells were incubated in the dark for 30 min at 4 °C and analyzed using a BD FACSLyric flow cytometer (BD Biosciences). Data were processed using BD FACSuite™ software v1.6 (BD Biosciences).

The initial quality assessment of the raw sequencing data was performed using FastQC (version 0.11.9) to evaluate the overall quality of the data. Low-quality bases and adapter sequences were trimmed using Trimmomatic (version 0.36) to ensure clean and high-quality reads. The cleaned reads were mapped to the reference genome (GRCm39) using HISAT2 (version 2.1.0), which is a highly efficient and fast alignment program. The mapping results were used for the subsequent quantification. Reads were quantified using RSEM (version 1.3.0), which provides accurate and reliable quantification of gene and isoform expression levels. Bowtie2 was used as part of the RSEM workflow for alignment. Differential expression analysis was conducted using the EdgeR program, with a significance threshold of P-value < 0.05, to identify DEGs. Gene sets were categorized based on GO terms and KEGG pathways. Enrichment analysis was performed using the enrichplot package (version 1.16.1), and the results were visualized using ggplot2 (version 3.3.6). GSEA was used to detect variations in signaling pathways between the high and low expression groups. Background gene sets were sourced from the Molecular Signatures Database (MsigDB) version 7.0. Gene set size setting: Analysis is performed with a minimum of 15 and a maximum of 500. Differential pathway expression analysis was conducted, and significantly enriched gene sets were identified based on consistency scores with an adjusted P-value < 0.05. The protein interaction network was constructed using the STRING database (https://string-db.org/) version 12.0 and visualized using Cytoscape (16). This network analysis provides insights into the functional interactions between proteins, enhancing our understanding of the molecular mechanisms involved.

All data are expressed as mean ± SD. Differences between the two groups were analyzed using Student’s t-test. Kaplan-Meier analysis was used to assess the differences between allograft survival curves and to calculate P-values. Statistically significance was set at P-value < 0.05. All statistical analyses were performed using GraphPad Prism version 10.4.1.

We established a mouse skin graft model to investigate the effects of amelogenin on graft rejection. A total 10 μg of recombinant mouse amelogenin (rM180) was applied to the dorsal skin (1.0 × 1.0 cm²) harvested from C57BL/6J donor mice and applied to the dorsal recipient bed of MHC-mismatched BALB/c recipient mice. The skin grafts were observed daily from days 7 to 14 post-transplantation and documented photographically. The syngeneic skin grafts satisfactorily adhered within two weeks ( Figure 1A , upper panel), whereas most of the skin grafts in the control group, to which phosphate-buffered saline (PBS) was applied, were significantly reduced in size at approximately day 12 and were eventually rejected ( Figure 1A , middle panel). Moreover, graft rejection in the rM180-applied group (rM180 group) was delayed by a median survival time of 5.5 days compared with that in the control group ( Figure 1A , lower panel, and Figure 1B ), and the necrotic area (black spots) of skin grafts in the rM180 group was consistently smaller than that in the control group from days 6 to 16 post-transplantation ( Figure 1C ).

To explore the effects of rM180 on the skin graft surroundings, inflammatory cell infiltration, and tissue damage were determined using histopathology of the skin grafts seven days after transplantation. The PBS-treated skin grafts showed histological signs of rejection and necrosis and were thick and swollen. Massive inflammatory cell infiltration was observed at the interface between donor and recipient skin. However, grafts in the rM180 group were thinner and had significantly fewer infiltrating cells ( Figure 2A ). The infiltration of T lymphocytes (CD4⁺, CD8⁺), B lymphocytes (CD19⁺), macrophages (F4/80⁺), and antigen-presenting cells (MHC II⁺) was determined by immunohistofluorescence analysis. In the grafts from the control group, clusters of CD4⁺ and CD8⁺ T cells ( Figure 2B ), CD19⁺ B cells ( Figure 2C ), F4/80⁺ macrophages ( Figure 2D ), and MHC II⁺ cells ( Figures 2C, D ) were detected seven days after transplantation. However, the frequencies of these clusters in grafts from the rM180 group were significantly lower ( Figures 2B–D ).

To study the effect of rM180’s potent inhibition of inflammatory cell infiltration in skin grafts on peripheral tissues, serum cytokine secretion levels were analyzed using flow cytometry. The concentration of cytokines peaked on day 7 post-transplantation, except for IL-2 and IL-6, which reached their peak on day 3 post-transplantation. In particular, the levels of the T-helper type 1 (Th1) cytokines IFN-γ ( Figure 3A ) and IL-2 ( Figure 3B ) in the serum of the rM180 group were much lower than those of the control group. On day 7 post-transplantation, the serum IFN-γ level in the rM180 group was only 30% of that in the control group, and on day 3 post-transplantation, the serum IL-2 level in the rM180 group was approximately 40% of that in the control group. This finding suggests the rM180-induced Th1 cell differentiation and dysfunction after allogeneic skin grafting. In contrast, no significant differences in the tumor necrosis factor-alpha (TNF-α) ( Figure 3C ) and IL-6 ( Figure 3D ) levels were observed between the two groups. In the rM180 group, the serum levels of IL-4 ( Figure 3E ) and IL-10 ( Figure 3F ) were slightly higher than those in the control group at all time points; however, no significant differences were observed. Furthermore, serum IL-17A levels in the rM180 group on day 7 after skin grafting were significantly lower than those in the control group ( Figure 3G ).

Because local administration of rM180 suppressed Th1 and Th17 cytokine levels in the peripheral blood, we investigated its effect on the spleen, a secondary lymphoid organ, after allogeneic skin transplantation. As shown in Figure 4A , splenic hypertrophy was observed in the control group on day 7 post-transplantation, whereas the spleens of the rM180 group exhibited a significant decrease in total weight and cell count. The spleens of the rM180 group were slightly larger than those in the wild type mice without skin transplantation ( Supplementary Figure 1 ). Flow cytometric analysis revealed lower percentages of CD4⁺ and CD19⁺ cells in the spleens of the rM180 group than in those of the control group ( Figures 4B, C ). Conversely, the two groups did not exhibit significant differences in the percentages of CD8⁺ and CD11b⁺ cells ( Figures 4B, D ). However, the cell count of each cell population in the spleen was significantly lower in the rM180 group than in the control group ( Figures 4B–D ). As shown in Figure 3 , IFN-γ, IL-2, and IL-17A concentrations in the peripheral blood of the rM180 group on day 7 post-transplantation were reduced. Therefore, we examined CD4⁺ cell subsets population in the spleen. The results revealed that the percentage of IFN-γ-positive cells in splenic CD4⁺ cells in the rM180 group was approximately 30% of that in the control group at 7 days post-transplantation ( Figure 4E ) and that the IL-17-positivity rate decreased to almost 50% ( Figure 4G ). In contrast, there were no differences in the percentage of IL-4 positivity in splenic CD4⁺ cells between the two groups ( Figure 4F ), and the percentage of splenic CD4⁺Foxp3⁺CD25⁺ cells in the rM180 group was almost twice that of splenic CD4⁺Foxp3⁺CD25⁺ cells in the control group ( Figure 4H ). These data suggest that local administration of rM180 resulted in decreased differentiation of CD4⁺ cells into Th1 and Th17 cells in the peripheral blood and increased differentiation into regulatory T (Treg) cells. This may contribute to the immune tolerance and delayed graft rejection by rM180 observed after skin transplantation.

As shown in Figure 3 , rM180 potently suppressed IL-2 production in peripheral blood from day 3 post-transplantation, while its systemic anti-inflammatory effects had largely disappeared after day 10. We therefore hypothesized that rM180 exerts an anti-inflammatory effect on the progression of inflammatory response that occurs immediately after transplantation. To investigate the potential molecular mechanisms of the rM180-induced prolongation of skin graft survival, RNA sequencing (RNA-seq) analysis of skin grafts from the rM180 group was performed on day 4 and 6 post-transplantation. Compared with the control group, 479 differentially expressed genes (DEGs), including 228 upregulated and 251 downregulated genes, were detected in the rM180 group on day 4 post-transplantation, and 302 DEGs, including 134 upregulated and 168 downregulated genes, were detected in the rM180 group on day 6 post-transplantation ( Figure 5A ). POU domain class 2 transcription factor 2 (Pou2f2) and lipocalin 2 (Lcn2) were among the genes that exhibited the most upregulated expression on day 4 post-transplantation, and chitinase-like protein 4 (Chil4) exhibited the most upregulated expression on day 6 post-transplantation (red box) ( Figures 5A, B ). In contrast, the expression of keratin-related genes (blue box) was highly downregulated on day 6 post-transplantation ( Figure 5B ). Inflammation and immune responses associated with skin graft rejection resulted in increased keratinization, suggesting that rM180 suppresses rejection-induced keratinization.

Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses were performed to further investigate the functions of the upregulated DEGs and the pathways involved in graft protection via rM180. The top 10 molecular functions are shown in Figure 5C . GO and KEGG analyses demonstrated that DEGs in the rM180-treated group were associated with “negative regulation of immune response” (red box) and “actin cytoskeleton reorganization” (blue box) on day 4 post-transplantation. On day 6, the genes that exhibited a change in expression levels in the rM180-applied group were mainly associated with “myocytes and myofibrils” (framed in green) ( Figure 5C ). Based on the above-mentioned analysis, the rM180-mediated prolongation of graft survival in skin graft models may be associated with the modulation of the inflammatory response within skin grafts and the regeneration of muscle fibers. The results of the KEGG analysis of inversely downregulated DEGs are shown in Supplementary Figure 2 .

To investigate the anti-inflammatory mechanism of rM180 in prolonging the survival of murine skin grafts, gene set enrichment analysis (GSEA) was performed. We demonstrated that T and B cell receptor signaling pathway, phagocytosis, natural killer cell-mediated cytotoxicity, toll-like receptor signaling pathway, Janus kinase/signal transducer and activator of transcription pathway and chemokine signaling pathway, apoptosis, and allograft rejection were downregulated on day 4 post-transplantation ( Supplementary Figure 3A ), whereas hedgehog signaling pathway and transforming growth factor β were upregulated at a higher level in the rM180 group than in the control group ( Supplementary Figure 3B ). Thus, GSEA on day 4 post-transplantation suggested that most pathways involved in rM180-mediated prolongation of skin graft survival were related to the regulation of the immune–inflammatory response. In contrast, the expression of genes associated with cardiomyopathy and myocarditis was downregulated on day 6 post-transplantation ( Supplementary Figure 3C ). This suggests that rM180 suppresses the destruction and degeneration of muscle fibers.

RNA-seq analysis demonstrated that Pou2f2 and Lcn2 were the most upregulated genes in the rM180 group 4 days post-transplantation ( Figure 5B ). Lcn2 is an acute-phase protein secreted by immune and epithelial cells in mucosal tissues. It is a 25-kD protein that covalently binds to matrix metalloproteinase-9 and is expressed in cells such as macrophages. In response to inflammation induced by various stimuli, Lcn2 levels increase and have been reported to mediate both pro- and anti-inflammatory responses (17). One study also has reported that Lcn2 promotes the process of skin wound healing in response to growth factors (18). First, immunostaining of grafts with Lcn2 demonstrated strong expression of Lcn2 in blood vessels or lymphatic vessels at the border between the rM180-applied skin graft and the recipient area ( Figure 6A ). To investigate whether the anti-inflammatory effects of amelogenin were caused by the Lcn2, anti-Lcn2 antibody, a neutralizing antibody against Lcn2, was injected around the graft every other day 24 h after amelogenin application. When the group injected with anti-Lcn2 antibody after application of rM180 was compared with the group injected with isotype control, hematoxylin and eosin (H&E) staining analysis 7 days after skin grafting confirmed that the grafts in the anti-Lcn2 group were thicker and had infiltrated immune cells compared with the isotype control group ( Figure 6B ). The size of the spleen in the rM180 group was increased by the anti-Lcn2 neutralizing antibody ( Supplementary Figure 4 ), and the decrease in the proportion of Th1 and Th17 cells in the spleen and the increase in Treg in the rM180 group was inhibited ( Figures 6C–E ). These results suggest that amelogenin-induced Lcn2 restricts the migration of inflammatory cells into the graft and reduces the proportion of Th1 and the Th17/Treg ratio in peripheral lymphoid tissues.

Pou2f2, also known as Oct2, is a B cell-regulatory transcription factor belonging to the POU domain family that uses the POU domain to bind to DNA (19, 20). Pou2f2 functions as a transcription factor that plays a pivotal role in B cell proliferation and differentiation by binding to the octamer DNA motifs present in the promoter of the immunoglobulin gene, and represses the expression of immunoglobulin in B cells (21). Next, we analyzed the distribution of Pou2f2 in skin grafts and observed a strong fluorescent signal of Pou2f2 in a band at the boundary between the graft and recipient bed in the rM180-applied group but not in the control group ( Figure 6F ). Furthermore, B cells in the skin grafts of the rM180 group strongly expressed Pou2f2 compared with those in the control group ( Figure 6G ).

Based on the results of the RNA-seq analysis demonstrating that Chil4 expression was the strongest on day 6 post-transplantation, immunostaining with Chil4 was performed on the grafts. The results demonstrated a strong fluorescent staining band of Chil4 at the graft recipient site in the rM180 group but not in the control group ( Figure 7A ). To investigate which proteins Chil4 interacts with and are involved in biological processes and signaling pathways, we performed protein–protein interaction analysis based on the results of RNA-seq analysis and identified nine hub genes, including upregulated genes such as Chil4, Chil3, resistin like alpha (Retnla), IL-4, IL-13, ribonuclease A family 2A (Rnase2a), chloride channel accessory 1 (Clca1), Mucin 5 subtype AC (Mus5ac), and Arginase 1 (Arg1) ( Figure 7B ). Macrophages are broadly classified into inflammation-induced M1 and wound-healing M2 cells. M1 macrophages are activated by lipopolysaccharides and other factors and produce proinflammatory factors, such as inducible nitric oxide synthase, TNF-α, IL-1β, and IL-6 (22, 23). In contrast, M2 macrophages are induced by IL-4 and IL-13, express CD206 and Arg1, and produce anti-inflammatory cytokines such as transforming growth factor-β, which are responsible for angiogenesis, removal of apoptotic cells, resolution of inflammation, and tissue repair (24–27). Thus, the macrophage is involved in both destruction and regeneration and plays an important role in the interface between inflammation and tissue regeneration. The M2 macrophage-related hub genes identified here are thought to be primarily involved in the negative regulation of the immune response, as demonstrated by the KEGG analysis ( Figure 5C ). As shown in Figure 7C , there was a strong expression of Arg1, an M2 macrophage marker, interspersed between rM180-applied grafts and the recipient bed, compared with the control group. Furthermore, flow cytometric analysis revealed that M2 macrophages significantly increased in skin grafts treated with rM180 ( Figure 7D ). These results suggest that M2 macrophages induced by rM180 may be responsible for tissue repair and graft protection at 6 d post-transplantation.