Type I interferon‐induced CD169⁺ macrophages play a protective role in periodontitis

Abstract

Background

This study aims to investigate the transcriptional characteristics and differentiation microenvironment of the CD169⁺ macrophages in periodontal tissues, and to explore their impact on periodontitis.

Methods

Single‐cell RNA sequencing landscape of periodontal tissues from healthy controls (n = 8) and periodontitis patients (n = 10) were constructed. Immunohistochemical staining was used to detect the percentage of CD169⁺ macrophages in periodontitis tissues of different severity to explore the potential links between them and periodontitis. Anti‐IFNAR1 antibodies were used to block the IFNAR1 receptor in mice to validate the role of type I interferon signaling in CD169⁺ macrophage differentiation and its impact on the progression of periodontitis. SCENIC analysis, immunohistochemical staining, immunofluorescence staining, flow cytometry, and in vitro experiments were used to identify the surface markers, differentiation microenvironment, and regulatory mechanisms of CD169⁺ macrophages.

Results

The phenotypes of CD169⁺ macrophages are characterized by CD169⁺, MERTK⁺, CX3CR1⁻, and CCR2⁻. The CD169⁺ macrophages could maintain immune homeostasis by producing CCL18 and IL10. The percentage of CD169⁺ cells in periodontal tissues was found to be negatively correlated with the severity of periodontitis. Type I interferons produced by gingival fibroblasts and keratinocytes are essential for the differentiation of CD169⁺ macrophages, which in turn orchestrate immunomodulation and tissue homeostasis restoration during the recovery phase of periodontitis.

Conclusions

Our findings indicate that type I interferons produced by fibroblasts in periodontal tissues induce monocytes to differentiate into macrophages with a CD169⁺MERTK⁺ phenotype. These cells play a role in negative immune regulation and may provide a positive contribution to the recovery of periodontitis.

Plain language summary

Periodontitis is one of the major diseases that threaten human health. CD169⁺ macrophages may play a key role in immune homeostasis by producing IL10 and CCL18. The type I interferon‐IRF7 axis may induce the differentiation of macrophages into CD169⁺ macrophages, which in turn orchestrates immunomodulation and tissue homeostasis restoration during the recovery phase of periodontitis. Studying the functions and differentiation mechanisms of CD169⁺ macrophages in periodontal tissue may offer new therapeutic targets for periodontitis and provide broader benefits for overall human health.

1. INTRODUCTION

Periodontitis was once considered as an inflammatory disease induced by microorganisms. However, it is now defined as a complex disease induced by not only periodontal pathogens but also host immune regulation. ¹ Therefore, the abundance and pathogenicity of periodontal bacteria may not be the main determinants. ² Current treatments, which primarily target biofilm removal, are sometimes less successful in improving the clinical outcomes for severe periodontitis (Stage III/IV) and preventing recurrence. ³ To achieve long‐term therapeutic goals, it is crucial to consider both the clinical parameters and the distinct characteristics of the host immune response. ⁴

Identifying immune cell populations that are important for maintaining host immune homeostasis could be valuable for the treatment of periodontitis. Periodontal tissues are frequently exposed to microbial stimuli, which contribute to a relatively complex immune microenvironment. ⁵ , ⁶ After being recruited to the periodontal tissues, immune cells (such as monocytes) may encounter a variety of different stimuli, which can lead to their differentiation into distinct functional clusters. CD169⁺ macrophages (also known as SIGLEC1⁺) are an increasingly studied subset recognized for their distinct roles in immune regulation, disease pathogenesis, and tissue repair. ⁷ , ⁸ Functionally, they help to limit tissue damage through pathogen clearance and modulation of inflammatory signaling. ⁷ , ⁹ CD169⁺ macrophages also facilitate regeneration by clearing apoptotic cells and enhancing osteoblast activity. ¹⁰ , ¹¹ , ¹² , ¹³ While prominently located in lymphoid organs such as lymph nodes and spleen, ¹⁴ CD169⁺ macrophages are also present in various peripheral tissues. ¹⁵ Currently, it remains unclear whether CD169⁺ macrophages exist within periodontal tissues, along with their biological characteristics and the microenvironment in which they differentiate.

Single‐cell RNA sequencing (scRNA‐seq) has revolutionized the ability to study the immune microenvironment of human peripheral tissues through detailed analysis of cell populations at single‐cell resolution. ¹⁶ In this study, we utilized scRNA‐seq to analyze comprehensively the myeloid cells in periodontal tissue. We referred to the approaches of studies in the field and merged our own data with two publicly available databases. ¹⁷ , ¹⁸ It was attempted to characterize the possible existence of CD169⁺ macrophages in periodontal tissues and to analyze their functions and clarify their possible involvements in periodontitis.

2. MATERIALS AND METHODS

2.1. Subjects

Human gingival tissues for scRNA‐seq analysis and histological staining were obtained from the Department of Periodontics, Stomatological Hospital of Tongji University. The study is in accordance with the Declaration of Helsinki, which was enacted in 1975 and revised in 2000. The study protocol has been approved by the Ethics Committee of Stomatology Hospital Affiliated to Tongji University (approval number: [2022]‐SR‐02). All participants provided written informed consent. The inclusion criteria for patients in this study were: (1) Patients diagnosed with periodontitis stage III or IV requiring periodontal surgery, or patients in good periodontal health requiring crown lengthening surgery. Among them, the severity of periodontitis and the healthy periodontal status were defined according to the new classification scheme for periodontal and peri‐implant diseases and conditions. ¹⁹ (2) Subjects without systemic diseases. The exclusion criteria were: (1) Participants who had taken anti‐inflammatory drugs or antibiotics in the past 6 weeks. (2) A former or current smoker (≥5 cigarettes per day). All participants received non‐surgical periodontal treatments and surgical treatments from the same experienced periodontist. The clinical information of all samples is shown in Supplementary table 1.

2.2. Animal experiments

Specific pathogen‐free C57BL/6 mice were purchased from Leagene Biotechnology. All animal feeding and experimental operations were conducted according to the Guide for the Care and Use of Laboratory Animals and the ARRIVE guidelines and were approved by the Ethics Committee of the Affiliated Stomatology Hospital of Tongji University (Approval No. [2022]‐DW‐07). All experiments used 8‐week‐old mice. All animals were maintained in specific pathogen‐free conditions. All data derived from animal studies were analyzed by an experimenter blind to experimental conditions. The sample size of each experiment was determined based on the effect size (assumed significance level at 0.05 and assumed power at 90%). The effect size was obtained according to standard deviation and mean difference of the pilot studies. The specific sample size of each experiment was described in the figure legends.

To establish a ligature‐induced periodontitis (LIP) model in mice, sterile 5.0 silk sutures were placed in the gingival sulci on both sides of the maxillary second molars and knotted on the palatal side for 8 or 16 days. Subsequently, models for the periodontitis phase and recovery phase were established (Figure 4K). The periodontitis phase model was generated by ligating the maxillary second molar for 8 days according to the above protocol. In other mice, the ligature was carefully removed to allow periodontal tissue healing for 8 days. Throughout the periodontitis phase and recovery phase, local injection of anti‐IFNAR1 antibody inhibited type I interferon signaling, thereby disrupting CD169⁺ macrophage differentiation. The effect of anti‐IFNAR1 antibody on suppressing the number of CD169⁺ macrophages is shown in the Figure S5A and S5B. Subsequently, gingival tissue or maxillary bone was harvested for subsequent experiments.

FIGURE 4.

In vitro experiments and LIP mouse experiments showed that the differentiation of CD169⁺ macrophages is dependent on IFN I signaling. (A–B) qPCR analysis of CD169 mRNA expression in THP‐1 cells (A) and BMDMs (B) under IFNα, IFNβ, IFNγ stimulation (20 ng/mL) for 4 h. Data were normalized to GAPDH and expressed as fold change and are represented as mean ± standard deviation of three biological replicates. (C–E) Representative flow cytometry images (C) and quantification results of CD169 expression in THP‐1 cells (D) and BMDMs (E). (F–G) Representative flow cytometry images (F) and quantification results (G) of CD169 expression in THP‐1 cells of in vitro co‐culture experiments. (H) Representative flow cytometry images of CD169⁺ macrophages (CD169⁺CD45⁺CD11b⁺) in gingival tissue of LIP mice at 8 and 16 days post‐ligature (n = 6–8 each). Gingival tissues from two mice were pooled together as one sample. (I–J) Quantification result of CD169⁺ macrophages (CD169⁺CD45⁺CD11b⁺) within the total macrophage population (CD45⁺CD11b⁺, I) and all immune cells (CD45⁺, J) in gingival tissue of LIP mice at 8 and 16days post‐ligature (n = 6–8 each). (K) Workflow for the in vivo model of the recovery phase of periodontitis: A ligature was placed on the maxillary second molar for 8 days, then carefully removed, allowing the periodontal tissues to heal for 8 days. Anti‐IFNAR1 was topically applied every other day throughout this period to block IFN I signaling. (L–M) Representative Micro‐CT reconstruction images (L) and quantitative analysis (M) of CEJ‐ABC distances and bone volume fraction (BV/TV %) of the maxilla in LIP mice between the iso group and anti group during the periodontitis phase (place ligature). Two white dashed lines delineate the maxillary molar CEJ line and the alveolar crest line, with the area between them representing bone resorption. The yellow straight line indicates the five measurement points for the CEJ‐ABC distance. (N–O) Representative Micro‐CT reconstruction images (N) and quantitative analysis (O) of buccal‐lingual CEJ‐ABC distances and bone volume fraction (BV/TV %) of the maxilla in LIP mice between the iso group and anti group during periodontitis recovery phase (remove ligature). Two white dashed lines delineate the maxillary molar CEJ line and the alveolar crest line, with the area between them representing bone resorption. The yellow straight line indicates the five measurement points for the CEJ‐ABC distance. (P) Representative images of H&E and Masson staining of the maxilla in LIP mice between the iso group and anti group (n = 5 each). The black dashed line indicates less alveolar bone repair in the anti group. The black dashed box shows that there is still a significant inflammatory cell infiltration in the periodontal tissues of the anti group. *p < 0.05, **p < 0.01, ***p < 0.001; HOKs, Human oral keratinocyte cell; HGFs, human gingival fibroblasts; LIP, ligature‐induced periodontitis. iso, isotype group; anti, anti‐IFNAR1 group; CEJ, cementoenamel junction; ABC, alveolar bone crest.

Anti‐IFNAR1 antibody (20 µg/mL ²⁰ ) was injected into gingival tissue using a microinjector (25 µL syringe with 33‐gauge needle; Hamilton) at 20 µL every 2 days for a total of 4 or 8 injections. This dosage and the local injection regimen were based on previous studies. ²¹ The isotype group and anti‐IFNAR1 group received equivalent volumes of isotype antibody, respectively.

2.3. Cell culture

The human leukemia cell line (THP‐1) and human oral keratinocyte cell line (HOKs) were cultured with Roswell Park Memorial Institute (RPMI) 1640 (Hyclone) with 10% fetal bovine serum (Excell Bio) and 1% penicillin/streptomycin solution (Gibco) in 5% CO² at 37°C. The macrophage differentiation of THP‐1 cells was induced by 100 ng/mL PMA (MCE) for 24 h. Bone marrow‐derived macrophages (BMDMs) were isolated by flushing the femurs and tibiae of 8‐week‐old C57BL/6J male mice. The BMDMs were incubated at 37°C for 7 days in Dulbecco's Modified Eagle's Medium (DMEM, Gibco) supplemented with M‐CSF (20 ng/mL, peprotech). Primary human gingival fibroblasts (HGFs) were derived, and cultures as described previously. Human gingival tissues were obtained during crown lengthening surgery, and tissues were cut into pieces of 1 mm² and cultured in DMEM supplemented with 10% FBS and 1% penicillin/streptomycin solution.

HOKs and HGFs were stimulated with lipopolysaccharide (LPS, InvivoGen) from Porphyromonas gingivalis (P.g) for 2/4/6 h for qPCR analysis. Differentiated THP‐1 cells and BMDMs were stimulated with human or mouse IFNα (peprotech), IFNβ (Biolegend), IFNγ (peprotech) for 4 h for qPCR analysis and 24 or 48 h for flow cytometry.

2.4. Single‐cell sequencing and downstream analysis

We collected gingival tissues from healthy individuals and patients with periodontitis and applied single‐cell sequencing. The inclusion and exclusion criteria were as described above. In order to expand the sample size of the present study, we merged our own data with information from publicly available databases (GSE171213 ²² and GSE152042 ²³ ).

Details of tissue processing and analysis are available in the Supplementary files

2.5. Flow cytometry and co‐cultures experiments

Alexa Fluor 647 anti‐human CD169 antibody (346005) was purchased from BioLegend. Gating strategies are shown in Figure S4. HOKs and HGFs were first stimulated with Pg.LPS for 24 h, the supernatants were collected and continued to be co‐cultured with THP‐1 with or without Anti‐IFNAR1 (A2460, Selleck).

Preparation of the gingival single‐cell suspension was performed according to previously described methods. ²⁴ , ²⁵ , ²⁶ The prepared single cells were incubated in the dark with fluorescence‐labeled antibodies for 20 min. The antibodies used in this study included anti‐mouse CD45‐FITC (103108), anti‐mouse CD11b‐APC‐Cy7 (101226), anti‐mouse CD169‐PE (142403), all purchased from BioLegend. BD FACS Verse TM System (BD Biosciences) and FlowJo software (FlowJoV10.0, TreeStar) were used for data acquisition and analysis.

2.6. Immunohistochemistry staining and immunofluorescence staining

The formalin‐fixed human gingival tissues were dehydrated and embedded in paraffin. Mice maxillae were decalcified in 10% EDTA (28 days) and embedded in paraffin. Tissues were cut into 5 µm thick sections from one end to another. The deparaffinized sections were subjected to antigen retrieval, sealed with 10% serum and 0.1% Triton X‐100 for 1 h, and then incubated overnight at 4°C with anti‐IFNα (DF6086, 1:100, Affinity), anti‐IFNβ (27506‐1‐AP, 1:100, Proteintech), anti‐CD169 (55427‐1‐AP, 1:100, Proteintech). The immunohistochemical staining KIT (KIT‐9706, MXB Biotechnologies) was used according to the manufacturer's instructions. Five positively stained sites were randomly selected and images captured under high‐power (400×) magnification. The average density of IFNα and IFNβ was measured by ImageJ, and the mean density of each site was calculated (Mean density = integrated optical density [IOD] sum/area). The number and percentage of CD169⁺ cells in the whole sample section were detected and calculated by positive cell detection function of Qupath (v0.5.0) software. ²⁷ , ²⁸

The tissue was processed as above and then incubated overnight at 4°C, using anti‐F4/80 (ab6640, 1:100, abcam), anti‐CD169 (ab312840, 1:100, abcam). After rinsing with PBS for 5 min three times, the tissue was incubated at room temperature with Donkey anti‐rat Alexa Fluor 488 or Donkey anti‐rabbit Alexa Fluor 594 for 1 h at room temperature. Finally, 4',6‐diamidino‐2‐phenylindole (DAPI) was used for nuclear staining at room temperature for 5 min. Images were taken using a Nikon fluorescence microscope (DS‐Ri1, Nikon).

2.7. Multiplex immunofluorescence staining

Multiplex immunofluorescence staining was performed using the TSA Detect Kit (Proteintech) following the manufacturer's protocol. After deparaffinization and rehydration, tissue sections underwent antigen retrieval and blocking. Primary antibodies of anti‐CD68 (66231‐2‐Ig, 1:1000, Proteintech), anti‐CD169 (55427‐1‐AP, 1:100, Proteintech), and anti‐MERTK (ab52968, 1:1000, abcam) were applied, followed by incubation with appropriate secondary antibodies. Staining was subsequently carried out using the fluorescent dyes (488, 594, and 647, Proteintech). Slides were scanned with a Pannoramic scanner (3DHISTECH) and images were analyzed using Slide Viewer software (3DHISTECH).

2.8. Quantitative PCR and siRNA transfection

Total RNA was extracted from cultured cells or human gingival tissue using TRIzol (Takara). Total RNA was subjected to reverse transcription using a Reverse Transcription kit (Takara). qPCR was performed utilizing the Hieff® qPCR SYBR Green Master Mix (Yeason) on the QuantStudio 5 Real‐Time PCR System (Thermo Fisher). We performed relative quantification using the 2^−ΔΔCt method with human GAPDH or mouse Gapdh as an internal control gene; data were expressed as relative fold changes. Primers used in the study could be found in Supplementary Table 5

Transfection was performed using RFectSP V2 (BIOG) following the manufacturer's protocol. Small interfering RNA (siRNA) was designed and synthesized by RIBOBIO. A negative control and siRNA targeting PRDM1 or IRF7 was used for knockdown experiments. To assess the efficiency of transfection, quantitative PCR (qPCR) was performed 48 h after transfection.

2.9. Statistical analysis

Data were presented as median (with interquartile range) or percentage for categorical variables and as mean (with standard deviation) for continuous variables. For continuous variables, the Shapiro‐Wilk test was first applied to examine for a normal distribution. The difference between two groups was analyzed by an independent t‐test if a variable was normally distributed or by the Mann‐Whitney test if otherwise. Multivariate comparisons were performed using one‐way ANOVA (Kruskal‐Wallis test for nonparametric test) with a Dunn's or Tukey multiple comparisons test. The Spearman correlation test was used for correlation analysis between clinical parameters and CD169⁺ cell percentage. A two‐tailed p < 0.05 was considered statistically significant. All statistical analyses were performed using GraphPad Software.

3. RESULTS

3.1. scRNA sequencing analysis reveals transcriptomic features of CD169⁺ macrophages in periodontal tissues

scRNA‐seq landscapes of periodontal tissues from healthy controls and periodontitis patients were constructed (Supplementary table 1). We merged our dataset with two public databases (GSE171213 ²² and GSE152042 ²³ ), obtaining a single‐cell transcriptome database of eight healthy controls and ten periodontitis patients (Figure 1A). After standard quality control, a total of 81,722 cells (34,961 from healthy controls and 46,761 from periodontitis patients) was obtained. All cells were clustered using uniform manifold approximation and projection (UMAP). Thirteen major clusters were finally identified (Figure 1B, S1). The periodontitis group contained an increased proportion of plasma cells relative to healthy gingiva (Figure 1B, Supplementary table 2). The volcano plot (Figure 1C) highlights the top five differentially expressed genes in each cluster.

FIGURE 1.

Overview of 81,722 cells of periodontal tissues from healthy controls (n = 8) and periodontitis patients (n = 10). (A) Workflow of this study. (B) Left panel: UMAP of 81,722 cells, colored by cell clusters. Middle panel: Stacked bar graph shows the ratio of different cell clusters among samples. Right panel: Proportion of each cell clusters in healthy control and periodontitis group. (C) The volcano plot shows the expression of classic marker genes in cell clusters, with boxes in the middle colored by cell cluster annotation. sigUP, significantly upregulated; sigDown, significantly downregulated.

Next, we merged and analyzed myeloid cells from three datasets, identifying 3692 cells across six clusters (Figure 2A, S2): two macrophage clusters (FCGR1A⁺CD68⁺, Macro), one monocyte cluster (CD14⁺CD16⁺), two dendritic cell clusters (ITGAX⁺MHC II⁺, DC), and one pre‐osteoclast cluster (CTSK⁺MMP9⁺, osteoclast precursors). The number of myeloid cells was higher in the periodontitis patients than the healthy controls (Figure 2B).

FIGURE 2.

Transcriptomic features of myeloid cells in periodontal tissue. (A) Left panel: UMAP of 3692 myeloid cells, colored by cell clusters. Right panel: Proportion of each cluster in healthy controls and periodontitis group. (B) Bar plot shows the mean number of myeloid cells in the two groups. (C) Stacked violin plot shows differential expression of markers in each myeloid cluster. (D–E) Pseudotime trajectory analysis reveals the differentiation trajectories of myeloid cells, consisting of a primary trajectory with two branches and a secondary trajectory. Colored by cell clusters. (F–H) Representative images (F) and quantitative analysis (G–H) of immunofluorescence staining of CD68⁺ cells and CD68⁺CD169⁺MERTK⁺ cells in healthy (n = 6) and periodontitis tissues (n = 6), scale bar = 100 µm. (I) GO:BP enrichment analysis of myeloid cell clusters. (J) Stacked violin plots showing comparative expression of CCL18, MRC1, CD163, and IL10 in each cluster between healthy control and periodontitis group. *p < 0.05, **p < 0.01, ***p < 0.001; cDC, Conventional dendritic cell; Pre‐osteoclasts, osteoclast precursors.

The first macrophage cluster highly expressed CD86 and IL1B and was identified as inflammatory macrophages (Inflammatory Macro). The second macrophage cluster exhibited distinct signatures with high expression of SIGLEC1 (CD169) and MERTK (Figure 2C). Meanwhile, this cluster also exhibited CX3CR1⁻ and CCR2⁻ phenotypes. This profile shared key features with the known transcriptional signatures of tissue‐resident macrophages in other tissues. ²⁹ , ³⁰ , ³¹ , ³² Given that CD169 is its most unique marker, we proposed that it might represent a population of CD169⁺ macrophages (CD169⁺ Macro) in periodontal tissues. In addition, the first dendritic cell cluster highly expressed XCR1, CLEC9A, and IRF8 (markers of cDC1). ³³ The second dendritic cell cluster, with high expression of CCR7, CD80, and LAMP3, was named CCR7⁺CD80⁺ DC. Finally, CTSK⁺MMP9⁺ cells were identified as pre‐osteoclasts.

To elucidate how myeloid cells differentiate under the microenvironment of periodontitis, we conducted single‐cell trajectory analysis using Monocle 2.0 to reconstruct the lineage from monocytes to macrophages. We found two major trajectories (Figure 2D,E). The primary trajectory started with monocytes, followed by Inflammatory Macro, which could finally differentiate into CD169⁺ Macro at the end of the trajectory. The secondary trajectory showed monocytes that could also directly differentiate into CD169⁺ Macro.

3.2. CD169⁺ macrophages are characterized by phagocytosis and secretion of CCL18

To address whether CD169⁺ macrophages exist in periodontal tissues, we collected gingival tissues from healthy controls and patients with periodontitis. Immunofluorescence staining detected the presence of cells co‐expressing CD68, CD169, and MERTK in human periodontal tissues (Figure 2F). We observed prominent infiltration of CD68⁺ macrophages in periodontitis tissues (Figure 2G). Although the number of CD68⁺CD169⁺MERTK⁺ cells also increased in periodontitis (Figure 2H), their proportion relative to the total macrophage pool may be reduced. Therefore, we proposed that their relative proportion, rather than the absolute number, more accurately reflected their association with periodontitis.

We then conducted Gene Ontology (GO) enrichment analysis on the five myeloid cell clusters (Figure 2I). The results showed that Inflammatory Macro had high expression of CCL3, S100A8, and S100A9, indicating their ability to recruit and activate neutrophils (Figure S3A). In contrast, CD169⁺ Macro may have strong phagocytosis capacity, which was shown by the higher expression of SIGLEC1 (CD169), C1QA, C1QB, and C1QC (Figure S3B,G). CD169⁺ Macro also showed expression of CCL18, a chemokine involved in regulatory T cell function. ³⁴ Notably, CCL18 levels in CD169⁺ Macro were significantly lower in periodontitis patients compared with healthy controls (P < 0.001, Figure 2J). Similarly, the levels of MRC1 (CD206) and CD163 in CD169⁺ Macro were significantly reduced in the periodontitis group (p < 0.001, Figure 2J). This implies that the functions of CD169⁺ Macro on maintaining immune homeostasis may be impaired during periodontitis.

Furthermore, monocytes exhibited high expression of VEGFA, a gene involved in angiogenesis, ³⁵ and were enriched in the “wound healing” GO term (Figure 2I, S3C). Both dendritic cell clusters had strong antigen presentation capabilities. However, CCR7⁺CD80⁺ DC possessed the capability to generate and control IL12 (Figure S3D–E), indicating their involvement in the maturation and differentiation of T cells. ³⁶ Finally, pre‐osteoclasts were associated with GO terms related to “bone resorption” and “osteoclast differentiation” (Figure S3F).

3.3. CD169⁺ cell percent is negatively correlated with periodontitis severity

As CD169 expression appears to be largely restricted to the macrophage populations, we sought to examine whether the abundance of CD169⁺ cells in periodontitis tissues correlates with disease severity. We collected a cohort of 29 periodontal tissues from patients diagnosed with periodontitis, excluding any patients who had factors that could affect the results. We performed immunohistochemical staining to quantify the number of CD169⁺ cells and used QuPath (v0.5.0) to calculate the number of positive cells in the whole tissue section.

A wide range of CD169⁺ cell numbers, varying from 56 to 1737, was observed (Figure 3A, Supplementary table 3). The results of the comparison between different severities of periodontitis showed that the probing depth (PD) < 5 mm group and the clinical attachment loss (CAL) < 5 mm group had a significantly greater number of CD169⁺ cells compared with the CAL > 9 mm group (P < 0.05, Figure 3B, Supplementary table 4). However, as periodontitis progresses, there might be a significant alteration in the quantity of immune cells that infiltrate the gingival tissues. We hypothesized that the percentage of CD169⁺ cells might be a more valuable measurement. As a result, the percentage of CD169⁺ cells varied between 0.097% and 1.608%. Furthermore, we observed a negative correlation between the percentage of CD169⁺ cells with PD, CAL, and the gingival index (GI, Figure 3C).

FIGURE 3.

CD169⁺ cell is negatively correlated with periodontitis severity. (A) Representative images of immunohistochemical staining of CD169⁺ cells in periodontitis patients with different severity (PD = 4,5,6 or 7 mm, n = 29 in total); The black dashed line delineates the epithelial layer and the lamina propria of the gingival tissue. (B) Scatter plots show the median of CD169⁺ cell numbers with interquartile range. (C) Spearman correlation analysis of PD, and CAL with CD169⁺ cell percentage (n = 29). (D) qPCR analysis for IFNA1 and IFNB1 expression of gingival tissues from healthy controls and periodontitis patients (n = 12 each). (E) qPCR analysis for IFNA1 and IFNB1 expression in HOKs and HGFs cells after stimulation with Pg.LPS (10 µg/mL) for 2, 4, and 6 h. Data were normalized to GAPDH and expressed as fold change and represent as mean ± standard deviation of three biological replicates. (F) Immunohistochemistry of IFNα and IFNβ in epithelium (middle) and the lamina propria (right) of gingival tissues from healthy controls and periodontitis patients (n = 5 each). Scale bar = 100 µm. (G) Quantitative analysis of mean density of IFNα and IFNβ in five randomized high‐magnification fields of view. Mean density = sum of integrated optical density (IOD)/area. *p < 0.05, **p < 0.01, ***p < 0.001; PD, probing depth; CAL, clinical attachment loss; GI, gingival index; PLI, plaque index.

3.4. Type I interferons from fibroblasts and keratinocytes may induce macrophage differentiation into CD169⁺ macrophages

Type I interferons (IFN Is, mainly IFNα and IFNβ) are well‐documented as stimulation for the expression of CD169 in macrophages. ¹³ , ³⁷ , ³⁸ To verify whether IFN Is are inducers of CD169⁺ macrophages, it was first necessary to examine the expression of IFN Is in periodontitis tissues. We examined the mRNA expression and protein level of the IFN Is in human gingival tissues. The results showed that the mRNA expression of IFNA1, IFNB1, and interferon‐stimulated genes, IFIT2, were significantly higher (p < 0.05) in periodontitis tissues (Figure 3D). Furthermore, in in vitro experiments, the qPCR results showed that the IFN Is gene was significantly upregulated in HOKs and HGFs after Pg.LPS stimulation (Figure 3E). Meanwhile, immunohistochemical staining also showed that the expression of IFNα and IFNβ was significantly higher in the epithelium and lamina propria of the gingival tissue in periodontitis (Figure 3F,G).

Subsequently, we stimulated THP‐1 differentiated macrophages and bone marrow‐derived macrophages (BMDMs) with IFN Is, as well as type II interferon IFNγ, for comparison. The expression of CD169 of macrophages was significantly upregulated under IFN Is stimulation (Figure 4A,B). Consistent results were obtained from flow cytometry after 24 and 48 h after IFN Is stimulation (Figure 4C–E). Co‐culture experiments were further used to investigate the mechanism of macrophage regulation in the microenvironment of periodontitis. We found that conditioned supernatants of HOKs/HGFs under Pg.LPS stimulation for 24 h also upregulate the expression of CD169 in macrophages. Moreover, anti‐IFNAR1 was applied to block the IFN Is signaling of macrophages and the results found that the induction of CD169 was suppressed accordingly (Figure 4F,G).

Furthermore, we investigated the dynamics of CD169⁺ macrophages in gingival tissue during periodontitis using a ligature‐induced periodontitis (LIP) mouse model. The gating strategy is detailed in Supplementary Figure S4. While the overall proportion of macrophages (CD45⁺CD11b⁺) in gingival tissue increased at 8 and 16 days post‐ligature (Figure 4H), the frequency of CD169⁺ macrophages (CD169⁺CD45⁺CD11b⁺) within both the total immune cell (CD45⁺) and macrophage populations was significantly reduced (Figure 4I,J). This specific decline in CD169⁺ macrophages suggest a potential disruption of local immune homeostasis during periodontitis.

To explore the role of IFN‐I signaling in this process, we administered a local injection of anti‐IFNAR1 antibody into the gingiva of LIP mice to block the IFN‐I signaling (Figure 4K). The number of CD169⁺ cells was assessed by flow cytometry in LIP mice 16 days post‐ligation. The results showed that blocking IFN I signaling significantly reduced the number of CD169⁺ cells in the gingival tissue (Figure S5A,B). Immunofluorescence staining also confirmed a significant decrease in the number of CD169⁺ macrophages in IFNAR1 blockade group (Figure S5C,D). MicroCT analysis of the mouse maxilla further revealed a significant reduction in the bone volume fraction (BV/TV%) in the anti‐IFNAR1 treatment group. However, the impact on buccal and palatal bone height was limited (Figure 4L,M). This limited effect on bone height could be attributed to the already low abundance of CD169⁺ macrophages during the periodontitis‐associated immune imbalance, potentially obscuring the full impact of their functional disruption.

Since we hypothesized that CD169⁺ macrophages are a population of long‐lived, reparative macrophages, and periodontitis involves a destructive‐reparative cycle. We then established a model of the recovery phase in periodontitis (Figure 4K). We removed ligatures on day 8 and allowed the periodontal tissue to heal for 8 days. Throughout the entire ligation and recovery phase, local injection of anti‐IFNAR1 antibody was administered to inhibit IFN‐I signaling, thereby interfering with CD169⁺ macrophage differentiation. MicroCT analysis (Figure 4N,O) revealed significantly less bone repair in the anti‐IFNAR1 (anti) group compared with the isotype (iso) group. Consistent with this, H&E and Masson staining of also confirmed diminished maxillary bone regeneration, reduced collagen repair, and less inflammatory cell infiltration within the periodontal tissues in the anti‐IFNAR1 group (Figure 4P). Taken together, these findings confirmed that IFN‐I signaling was essential for CD169⁺ macrophage differentiation and their critical role in bone recovery during the recovery phase of periodontitis.

3.5. IFN Is could induce CD169⁺ macrophage markers and promote CCL18 and IL10 secretion

To further characterize the regulatory effects of IFN Is on macrophages, we examined the expression of additional surface markers. In THP‐1‐derived macrophages, stimulation with IFN Is significantly increased the expression of CD80, CD86, CD206, CD163, MERTK, and MHC class II molecules (Figure 5A,B). Notably, we also observed a marked upregulation of CX3CR1 and CCR2 (Figure 5B), which contrasts with the expected phenotypic profile of CD169⁺ macrophage.

FIGURE 5.

SCENIC analysis reveals that IRF7 may regulate IFN Is‐induced macrophage upregulation of CD169. (A–C) qPCR analysis of CD169⁺ macrophage markers (MERTK, HLA‐DRA, CD206, CD163, CD86, CD80, CX3CR1, CCR2, IL1B, TNFA, CCL18, IL10) in THP‐1 cells. (D–F) qPCR analysis of CD169⁺ macrophage markers (Mertk, H2aa, Cd206, Cd163, Cd86, Cd80, Cx3cr1, Ccr2, Il1b, Tnfa, Il10) in BMDMs. Data were normalized to GAPDH and expressed as fold change and are represented as mean ± standard deviation of three biological replicates. (G–H) qPCR analysis of CX3CR1, CCR2 in THP‐1 cells and BMDMs. Data were normalized to GAPDH and are expressed as fold change and are represented as mean ± standard deviation of three biological replicates. (I) Heatmap showing the activity of transcription factors among different myeloid cell clusters, activity scores were predicted by SCENIC analysis. High‐activity transcription factors was highlighted and colored by cell clusters. (J) Efficiency of siRNA knockdown of IRF7 in THP‐1 cells. (K) Expression of transcription factors and IRF7 in THP‐1 cells under stimulation of IFNα, IFNβ. (L) Expression of CD169 in THP‐1 cells under stimulation of IFNα, IFNβ with siRNA knockdown of IRF7. *p < 0.05, **p < 0.01, ***p < 0.001.

In contrast, the response of BMDMs to IFN Is differed in several aspects. While IFN Is only weakly induced the expression of Mertk and Cd206, they significantly downregulated Cd163 (Figure 5D). Furthermore, although Cd86 expression was markedly enhanced by IFN Is, no significant change was detected in Cd80 levels. (Figure 5E). Finally, IFN Is stimulation led to a pronounced decrease in Cx3cr1 expression in BMDMs (Figure 5E).

These discrepancies suggest that IFN Is alone is insufficient to recapitulate the full phenotypic signature of CD169⁺ macrophages, whose differentiation is likely shaped by a combination of signals within the immune microenvironment. Previous studies have shown that LPS can negatively regulate CCR2 expression in macrophages. ³⁹ To explore this in our in vitro model, we treated both THP‐1‐derived macrophages and BMDMs with LPS and IFNβ, either individually or in combination. We found that dual stimulation significantly downregulated CX3CR1 and CCR2 mRNA levels in both macrophage types (Figure 5G,H), providing a partial explanation for the phenotypic regulation observed in CD169⁺ macrophages.

Furthermore, the expression of IL1B, TNFA, CCL18, and IL10 was also analyzed. Our findings indicated that interferons increased the expression of IL1B (Figure 5C–F). However, IFN Is did not increase the expression of TNFA, but IFNγ did (Figure 5C). Nevertheless, the expression of Tnfα in BMDMs was significantly upregulated, but it was not in IL1β (Figure 5F). Notably, IFN Is could upregulate the expression of CCL18 in THP‐1 differentiated macrophages, and IL10 in both THP‐1 differentiated macrophages and BMDMs (Figure 5C–F).

3.6. SCENIC analysis reveals that IRF7 may regulate IFN Is‐induced macrophage upregulation of CD169

Next, we applied SCENIC to evaluate the transcription factor activity of myeloid cell clusters with the aim of identifying the transcription factors that regulate the phenotype of CD169⁺ Macro. We observed that CD169⁺ Macro exhibited high activity of PRDM1. Furthermore, as we found that monocytes may be a precursor to CD169⁺ Macro in the above results, the transcription factor activity of monocytes also needs to be considered. It was discovered that monocytes showed high activity in IRF7, which was associated with IFN Is‐related pathways (Figure 5I).

We analyzed the expression of transcription factors in macrophages that were activated by IFN Is using qPCR. PRDM1 and IRF7 showed significant upregulation in macrophages (Figure 5K, S6C). Next, siRNA transfection was used to knock down these two transcription factors to investigate their regulatory functions on CD169. The data presented showed the efficiency of siRNA transfection and knockdown in THP‐1 cells (Figure 5J, S6A–B). The knockdown of IRF7 led to a significant downregulation in the expression of CD169 in macrophages (Figure 5L).

4. DISCUSSION

Monocytes, macrophages, dendritic cells, and osteoclasts are a closely related family of myeloid cells characterized by their crucial roles in pathogenic microbial recognition and immunomodulatory. ²² , ⁴⁰ Periodontal tissues are continuously stimulated by external microorganisms, and the immune microenvironment is therefore quite complex. ⁵ , ⁶ Upon recruitment into periodontal tissues, monocytes may encounter a variety of different stimuli, thereby differentiating into distinct functional clusters.

An increasing number of studies are questioning the conventional binary categorization of M1/M2 type macrophages. ⁴¹ , ⁴² Here, we detected a population of macrophages with a CD169⁺MERTK⁺ phenotype within periodontal tissues. We speculated that they may be, and refer to them as, CD169⁺ macrophages and explored the potential association between them and periodontitis through in vivo and in vitro experiments. Analysis of single‐cell sequencing data revealed that the phenotypic characteristics of CD169⁺ Macro are CD169⁺, MERTK⁺, CX3CR1⁻, and CCR2⁻. Meanwhile, CD169⁺ Macro's functions and their disease impacts resemble those of tissue resident macrophages or CD169⁺ macrophages in other tissues. ²⁹ , ⁴³ , ⁴⁴ , ⁴⁵ Overall, CD169⁺ macrophages may possess anti‐inflammatory regulatory functions, such as removing apoptotic cells and debris, promoting tissue repair, and producing anti‐inflammatory cytokines IL10 and CCL18. ⁴⁶

We then evaluated the potential association of CD169⁺ macrophages with the severity of periodontitis. Our analysis of the periodontitis samples revealed a negative correlation between the proportion of CD169⁺ macrophages and disease severity. During peak inflammation in LIP mice, these cells constituted a relatively low proportion (about 0.5%) of CD45⁺ cells. This low abundance likely resulted from the concurrent massive infiltration of neutrophils and other immune cells. We then hypothesize that CD169⁺ macrophages play a more significant role during the recovery phase of periodontitis. We then tested this hypothesis using a recovery model (8‐day ligature followed by 8‐day healing) and observed that local IFNAR1 blockade significantly impaired alveolar bone repair. These results not only confirm the essential role of type I interferon signaling in the functional differentiation of CD169⁺ macrophages, but also indicate the positive contribution of CD169⁺ macrophages to tissue repair during the recovery phase of periodontitis. Collectively, these findings suggest that CD169⁺ macrophages may play an essential immunomodulatory role in restoring tissue homeostasis, particularly during the recovery phase of periodontitis.

We then identified IFN‐Is, likely derived from periodontal fibroblasts and keratinocytes, as a key inducer of the CD169⁺ macrophage phenotype. This was confirmed in co‐culture models and LIP mice. Consistent with this, IFN‐I upregulated CD169⁺ macrophage markers (CD169, MERTK, MHC II, IL10, and the CCL18) in macrophages. However, the IFN Is signaling does not fully explain the unique phenotype of CD169⁺ macrophage. The transcription of CX3CR1 and CCR2 appears to contradict our initial in vitro findings that IFN‐I alone can upregulate these receptors. However, subsequent experiments revealed that the co‐stimulation with LPS, a key pathogen‐associated signal in periodontitis, significantly suppressed the IFN‐I‐induced expression of CX3CR1 and CCR2. This observation helps to reconcile the apparent paradox by suggesting that the definitive CD169⁺ macrophage signature observed in vivo is probably not driven by IFN‐I alone. Instead, this process might involve input from other signals in the inflammatory microenvironment, such as those derived from microbes.

We further employed SCENIC transcription factor analysis to determine the transcription factor activity in different myeloid clusters. Based on the previous studies, we speculated that highly activity transcription factors, IRF7 and PRDM1, may play regulatory roles in the transcription of CD169. We confirmed that IRF7 mRNA expression is significantly upregulated under IFN‐Is stimulation. However, the knock down of IRF7 significantly inhibited the expression of CD169 in macrophages induced by IFN‐Is. These findings may elucidate the regulatory role of IRF7 in the expression of CD169 in macrophages.

This study has several limitations that should be considered. First, our experimental validation relied on bone marrow/monocyte‐derived in vitro models. Future studies should employ more definitive approaches, such as lineage tracing or long‐term in situ fate‐mapping, to rigorously determine the origin and maintenance of these CD169⁺ macrophages within periodontal tissues. While we establish IFN‐I signaling as necessary for CD169⁺ macrophage differentiation, the complexity of the inflammatory microenvironment nevertheless suggests the involvement of additional pathways, potentially including those triggered by bacterial components such as LPS. Future spatial transcriptomic approaches ⁴⁷ will help to dissect this multicellular signaling network. Moreover, the clinical correlation analysis between CD169⁺ macrophages and periodontitis progression were constrained by a relatively small number of tissue samples from both mild (<5mm) and severe (>9mm) cases, which may limit the statistical power and introduce potential bias. Finally, although locally administered, the IFNAR1‐blocking antibody could affect other IFNAR1⁺ cells in the tissue, a potential confounder that remains difficult to exclude.

5. CONCLUSION

Overall, our findings indicated that type I interferons produced by fibroblasts in periodontal tissues induce monocytes to differentiate into macrophages with a CD169⁺MERTK⁺ phenotype. These cells play a role in negative immune regulation and may provide a positive contribution to the recovery of periodontitis.

AUTHOR CONTRIBUTIONS

Jiawei Lu and Zehui Xiong: Contributed to conception; design; data acquisition and interpretation; drafted and critically revised the manuscript. Xiao Wu, Yanan Yang and Jinyi Zhang: Contributed to data acquisition and performed all statistical analyses; drafted, and critically revised the manuscript. Lijun Luo: Contributed to conception; design; data interpretation, and critically revised the manuscript.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflict of interest. All authors gave their final approval and agreed to be accountable for all aspects of the work.

Supporting information

Supporting Information

DATA AVAILABILITY STATEMENT

The data used and/or analyzed during the current study are available from the corresponding author on reasonable request. This study utilized two publicly available single‐cell datasets from the GEO database, GSE171213 (DOI: 10.7150/thno.65694) and GSE152042 (DOI: 10.7554/eLife.62810).