The THP-1 human monocytic cell line was purchased from Biospecies (Guangdong, China). THP-1 cells were cultured in Roswell Park Memorial Institute 1640 (RPMI 1640, Procell, Wuhan, China) medium supplemented with 10% fetal bovine serum (FBS, PAN, Aidenbach, Germany) and 1% penicillin-streptomycin (Beyotime, Shanghai, China). To induce differentiation into M1 macrophages, THP-1 cells were stimulated with 200 ng/ml phorbol 12-myristate 13-acetate (PMA, Sigma, Shanghai, China) for 24 h to form adherent M0 macrophages, followed by treatment with 100 ng/ml LPS and 20 ng/ml IFN-γ for 48 h.

Human mesenchymal stem cells were obtained from Oricell (Guangzhou, China) and cultured in Minimum Essential Medium α (α-MEM, Procell, Wuhan, China) supplemented with 10% FBS and 1% penicillin-streptomycin.

All cell cultures were maintained in Esco CelMate carbon dioxide (CO₂) incubators (Esco, Singapore) at 37 °C under 5% CO₂ conditions to simulate physiological environments.

Small interfering RNAs (siRNAs) specifically targeting NRIR and RSAD2 were purchased from OBIO (Shanghai, China), with non-targeted siRNA used as negative control (NC). The siRNA and NC sequences are provided in Supplementary Table S1 . Overexpression plasmids containing full-length RSAD2 and control plasmids were also acquired from OBIO (Shanghai, China); plasmid details are shown in Supplementary Figure S1 .

For siRNA transfection, 5 × 10⁵ THP-1 cells were seeded into six-well plates and cultured for 24 h, followed by incubation in Opti-MEMI Reduced Serum Medium (31985-070, Gibco, California, USA) for an additional 12 h. Subsequently, 100 pmol of siRNA was transfected into each well using 7.0 μl CALNPTM RNAi reagent A and 2.0 μl CALNPTM RNAi reagent B (D-Nano, Beijing, China).

For plasmid transfection, 5 × 10⁵ THP-1 cells were seeded into six-well plates for 24 h. Upon reaching approximately 80% confluency, 3 μg of plasmid DNA was transfected per well with 4.8 μl Lipofect5000 reagent and 200 μl Trans buffer (BIOG, Changzhou, China).

Total RNA was isolated from M1 macrophages using TRIzol reagent (Sangon, Shanghai, China) following the manufacturer’s instructions. RNA quality assessment, library preparation, sequencing, quality control, mapping reads to the reference genome, and differential gene expression analyses were performed by Novogene Co., Ltd (Beijing, China). For functional annotation, Gene Ontology (GO, http://www.geneontology.org/) and Kyoto Encyclopedia of Genes and Genomes (KEGG, http://www.genome.jp/kegg/) analyses were conducted to elucidate gene functions and relevant pathways associated with differentially expressed genes (DEGs). Gene Set Enrichment Analysis (GSEA) was further performed to identify key KEGG signaling pathways. R software (version 4.40) was utilized to visually represent data through principal component analysis (PCA), volcano plots, heatmaps, and pathway enrichment plots. The RNA-seq data were deposited in the NCBI Sequence Read Archive (SRA) (https://www.ncbi.nlm.nih.gov/sra) (BioProject ID: PRJNA1281584; BioSample accession numbers: SAMN49569226, SAMN49569227, SAMN49569228, SAMN49569229, SAMN49569230, and SAMN49569231).

Total RNA was extracted from THP-1 macrophages and BMSCs using the SteadyPure Rapid RNA Extraction Kit (Accurate Biology, Hunan, China). RNA was reverse-transcribed into complementary DNA (cDNA) using the Evo M-MLV RT Mix Kit (Accurate Biology, Hunan, China). Subsequently, RT-qPCR was performed using SYBR^® Green Premix Pro Taq HS RT-qPCR Kit (Accurate biology, Hunan, China), following the manufacturer’s instructions. Relative gene expression levels were calculated using the 2^-ΔΔCt method as described by Livak et al. (21). Primers synthesized by Sangon Biotech (Shanghai, China) are listed in Supplementary Table S2 .

Total proteins were extracted from THP-1 cells and BMSCs using a total protein extraction kit (Keygen Biotech, Jiangsu, China). Protein concentrations were quantified using a BCA assay Kit (Beyotime, Beijing, China). After centrifugation (12,000 × g, 4°C, 10 min), protein lysates were separated by 10% or 12.5% SDS-PAGE and transferred onto polyvinylidene fluoride (PVDF) membranes. Membranes were blocked with 5% skimmed milk in Tris-buffered saline containing Tween-20 (TBST) at room temperature for 2 h, then incubated overnight (4°C) with primary antibodies against RSAD2 (Proteintech, 28089-1-AP, 1:1000), CD86 (Huabio, ET1606-50, 1:1000), iNOS (Huabio, HA722031, 1:1000), phospho-p65 (CST, 3033 T, 1:1000), p65 (Proteintech, 10745–1-AP, 1:1000), phospho-IκB (Huabio, HA722770, 1:1000), IκB (Huabio, ET1603-6, 1:1000), Osteopontin (OPN, Proteintech, 80912-4-RR, 1:4000), osteocalcin (OCN, Bioss, bs-4917R, 1:1000), runt-related transcription factor-2 (RUNX2, Huabio, ET1612-47, 1:5000), and GAPDH (Affinity, AF7021, 1:1000). Subsequently, membranes were washed three times in TBST (10 min each) and incubated with HRP-conjugated secondary antibodies (Goat anti-Rabbit IgG, Proteintech, SA00001-2, 1:4000) at room temperature for 2 h. Protein bands were visualized using enhanced chemiluminescence reagent (Affinity, West Virginia, USA) and an iBright CL1000 imaging system (Thermo, MA, USA). Semi-quantitative analysis of protein expression was performed using ImageJ software (v1.8.0, NIH, MD, USA).

Following successful macrophage polarization, THP-1 cells were harvested and filtered to prepare single-cell suspensions. Fc receptor-mediated nonspecific binding was blocked using human Fc receptor blocking solution (TruStain FcX™, BioLegend, 422301). Cells were washed three times, stained on ice with PE-conjugated anti-CD86 antibody (Biolegend, 374205) in staining buffer containing 1% FBS, and analyzed using a FACSMelody flow cytometer (BD). Data analysis was performed using FlowJo software (Tree Star).

THP-1 cells were fixed in 4% paraformaldehyde (PFA) for 30 min, permeabilized with 0.5% Triton X-100 for 30 min at room temperature (20°C), and blocked with 5% goat serum in phosphate-buffered saline (PBS) for 1.5 h (37°C). Subsequently, cells were incubated overnight (4°C) with primary antibodies against CD86 (Huabio, ET1606-50, 1:500) and iNOS (Huabio, HA722031, 1:100). After three washes with PBS, cells were incubated with fluorescent secondary antibodies for 1 h (37°C, in darkness). Cell nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI, Solarbio, Beijing, China). Stained cells were visualized using laser scanning confocal microscopy (BC43 SR, OXFORD, UK).

M0 macrophages were transfected and subsequently polarized into M1 macrophages using LPS/IFN-γ for 48 h. Cells were then washed three times with PBS and cultured in serum-free medium for an additional 24 h. SNs were collected, centrifuged (300 × g, 10 min) to remove cellular debris, and filtered (0.22 μm filter). A portion of SNs was combined 1:1 with osteogenic induction medium (Oricell, Guangzhou, China) for BMSC co-culture experiments, and the remainder was used directly for in vivo experiments.

BMSCs underwent osteogenic induction upon reaching 80% confluency. After 14 days, ALP activity was assessed using an ALP staining Kit (Beyotime, Beijing, China). Calcium mineralization was evaluated after 21 days by Alizarin Red S (ARS) staining (Oricell, Guangzhou, China). Chromogenic reactions were visualized using a stereomicroscope (SZN71, SOPTOP, China).

Thirty male Sprague-Dawley (SD) rats (8 weeks old) were obtained from the Animal Center of Southwest Medical University. All surgical procedures were approved by the Animal Ethics Committee of Southwest Medical University (SWMU20210414) and followed the ARRIVE (Animal Research: Reporting of In Vivo Experiments) guidelines.

Detailed surgical procedures were described in our previous studies (22). Briefly, unilateral maxillary first molars of SD rats were extracted under general anesthesia induced by 4% isoflurane inhalation. After a 4-week healing period, rats were anesthetized similarly, and local anesthesia (articaine with 1:100,000 epinephrine; Primacaine, France) was administered at the surgical site. Gingival incisions were made in the maxillary first molar region, and implant sockets were prepared using a reamer (1.6 mm diameter). Customized Ti-6AL-4V screw-type implants (2mm diameter, 3mm thread length, 1.5mm smooth neck) (22) were then inserted. Four weeks were allowed for osseointegration. Seven rats were excluded after four weeks due to death (n = 3) or implant loosening (n = 4). The remaining rats were randomly allocated into four groups using a random number table: (a) control group (n = 5); (b) LPS group (n = 6); (c) si-NC-SN group (n = 6); and (d) si-NRIR-SN group (n = 6).

Different treatments were applied to each group. In the LPS group, LPS derived from P. gingivalis (1 mg/mL, 10 µL per injection) was injected into the gingival sulcus around implants every three days for two weeks to establish peri-implantitis (23). PBS replaced LPS injections in the control group. Considering rats lack the NRIR gene, direct knockdown via vectors (such as nanoparticle- or adenovirus-mediated siRNA transfer) was infeasible. Referring to prior studies (Tahmasebi et al., Ma et al.) (24–26), we chose macrophage-secreted supernatants (SN) as media to investigate NRIR’s role in animal models. Rats in the si-NC-SN group received SN from M1 macrophages transfected with non-targeting siRNA, while those in the si-NRIR-SN group received SN from NRIR siRNA-transfected M1 macrophages. SN (200 µL) was injected into both buccal and lingual gingiva around implants every 3 days for 2 weeks. Rats were sacrificed four weeks post-injection, and tissues were harvested for analysis.

Maxillae were scanned using micro-CT (Inveon PET CT, Siemens, Germany) with the following parameters: spot size (50 µm), tube voltage (80 kVp), tube current (500 µA), and total rotation (360°). Three-dimensional (3D) reconstructions were analyzed using Mimics (version 21.0) software. Peri-implant bone loss was assessed by measuring the distance from the most coronal marginal bone position to the apical implant head at distal, mesial, buccal, and palatal locations. Bone mineral density (BMD), trabecular thickness (Tb.Th), bone volume fraction (BV/TV), trabecular separation (Tb.Sp), and trabecular number (Tb.N) were quantified according to the micro-CT analysis guidelines outlined by Huang et al. (27).

Maxillary alveolar bone specimens were fixed in 4% paraformaldehyde and decalcified for 6 weeks 10% ethylene diamine tetraacetic acid (EDTA, Solarbio, Beijing, China). Subsequently, specimens were sectioned into 4 µm slices (proximal-distal direction), blocked with 5% BSA for 1h, and incubated with primary antibodies overnight (4°C) according to manufacturer protocols. Slices were then incubated with DAB chromogenic solution, counterstained with hematoxylin (3 min), and visualized under an Olympus BX51 microscope (Tokyo, Japan). Semi-quantitative analysis was conducted using ImageJ software (v1.8.0, NIH, MD, USA). Staining intensity was quantified by calculating average optical density (AOD) as the ratio of integrated optical density (IntDen) of positively stained regions to the area (AOD = IntDen/Area).

Statistical significance was assessed using two-tailed Student’s t-test or one-way ANOVA. Where ANOVA showed significant differences, Student-Newman-Keuls q (SNK-q) post-hoc tests were performed to identify statistically significant pairwise differences. Variables not meeting normal distribution criteria were analyzed using the Kruskal–Wallis H test. Statistical analyses were performed using SPSS software (version 27.0, SPSS Inc., Chicago, USA). Quantitative data are presented as mean ± standard deviation (SD). Differences were considered statistically significant at P < 0.05.

In our previous study, high-throughput transcriptome sequencing identified lncRNA NRIR in gingival tissues from peri-implantitis patients. NRIR may critically contribute to peri-implant inflammation (20). Additionally, seven genes (RSAD2, CMPK2, IFIT1, IFIT3, ISG15, BST2, and HLA-C) were predicted as potential NRIR targets (20).

Since macrophages are pivotal immune cells in peri-implantitis, we hypothesized that NRIR might regulate macrophage polarization. To test this hypothesis, we analyzed M1 macrophages and observed that NRIR was significantly upregulated ( Figure 1A ). Next, we used specific siRNA to silence NRIR and evaluated its influence on M1 macrophage activation. RT-qPCR results indicated significant downregulation of M1-associated genes (IL-1β, IL-6, iNOS, CD86, and CXCL10) following NRIR knockdown ( Figure 1B ). Western blotting demonstrated reduced CD86 protein expression in NRIR-knockdown macrophages ( Figure 1C ). Flow cytometry analysis revealed that NRIR knockdown significantly decreased M1 polarization, as indicated by CD86 ( Figure 1D ). Additionally, IF assays confirmed that NRIR silencing markedly inhibited expression of M1 macrophage polarization markers (iNOS, CD86) ( Figure 1E ).

To further explore the mechanisms underlying NRIR-mediated macrophage activation, we performed transcriptome microarray and bioinformatics analyses on M1 macrophages after NRIR knockdown. PCA and heatmap analyses demonstrated distinct transcriptomic profiles between control and NRIR-knockdown groups, indicating significant gene expression alterations ( Figures 2A, B ). Differential expression analysis identified 729 DEGs (357 downregulated, 372 upregulated), visualized by a volcano plot ( Figure 2C ). GSEA enrichment indicated general downregulation of the NF-κB signaling pathway in NRIR-knockdown macrophages ( Figure 2D ). Additionally, intersection analysis between DEGs and predicted NRIR targets identified RSAD2 as the only overlapping gene ( Figure 2E ). Preliminary characterization demonstrated that RSAD2 was downregulated following NRIR knockdown at both mRNA and protein levels, whereas RSAD2 knockdown did not affect NRIR expression ( Figures 2F, G ). Thus, we strongly suspected RSAD2 as a potential target of NRIR in macrophage polarization regulation.

To investigate the role of RSAD2 in M1 macrophages, RSAD2-knockdown macrophages were evaluated. At the genetic level, M1-associated genes (IL-1β, IL-6, iNOS, CD86, and CXCL10) were significantly downregulated ( Figure 3A ). At the protein level, RSAD2-knockdown macrophages showed reduced expression of RSAD2 and CD86 ( Figure 3B ). Flow cytometry results demonstrated that RSAD2 knockdown markedly reduced M1 polarization ( Figure 3C ). Moreover, IF confirmed that RSAD2 silencing effectively suppressed the expression of M1 macrophage polarization markers (iNOS, CD86) ( Figure 3D ).

To confirm whether NRIR exerts its effects by regulating RSAD2, a rescue assay was conducted. Plasmids targeting RSAD2 were transfected into THP-1 macrophages with confirmed NRIR knockdown to restore RSAD2 expression. Results indicated that RSAD2 overexpression reversed the reduced gene and protein expression associated with M1 macrophages caused by NRIR knockdown ( Figures 4A, B ).

We subsequently investigated signaling pathways regulated by NRIR during macrophage polarization. GSEA indicated that NRIR knockdown inhibited the NF-κB signaling pathway ( Figure 2D ). The NF-κB pathway is essential for M1 macrophage activation, particularly in regulating inflammatory gene expression (28). Western blot analysis assessed phosphorylated forms of key factors, IκB and p65. Results showed that NRIR knockdown reduced phosphorylated IκB, phosphorylated p65, and CD86 levels in M1 macrophages. However, this reduction was reversed by treating cells with Diprovocim, a potent TLR1/TLR2 agonist that activates NF-κB signaling downstream ( Figure 5A ).

To confirm whether NRIR regulates NF-κB signaling via RSAD2, we evaluated RSAD2-knockdown macrophages. Protein levels of phosphorylated IκB, phosphorylated p65, and CD86 decreased in RSAD2-knockdown macrophages during M1 activation. These findings indicate that NRIR depletion suppresses NF-kB signaling by downregulating RSAD2, further attenuating M1 macrophage polarization ( Figure 5B ).

Chemokines and cytokines secreted by macrophages regulate MSC migration and differentiation for bone regeneration (29–31). To evaluate effects of factors secreted by NRIR-knockdown macrophages on BMSC osteogenic differentiation, conditioned medium from NRIR-knockdown M1 macrophages was used to culture BMSCs. RT-qPCR and Western blotting showed that NRIR-knockdown macrophage supernatants significantly increased expression of osteogenic differentiation markers (OCN, OPN, RUNX2) in BMSCs compared to control groups ( Figures 6A, B ). ALP staining demonstrated enhanced ALP expression in BMSCs cultured with NRIR-knockdown macrophage supernatants ( Figure 6C ). ARS staining showed that BMSCs cultured with NRIR-knockdown macrophage supernatants formed more mineralized nodules ( Figure 6D ).

Next, we explored the role of NRIR in peri-implantitis using a rat model. Clinical observations showed that supernatants from macrophages transfected with si-NC produced clinical manifestations similar to LPS-induced peri-implantitis. However, inflammation was significantly lower in rats treated with supernatants from macrophages transfected with si-NRIR ( Figure 7A ). Micro-CT revealed that bone loss in the si-NC-SN group and LPS group was comparable, whereas bone resorption was significantly lower in the si-NRIR group ( Figure 7B ). Bone morphometric analysis demonstrated significantly increased BMD, BV/TV, Tb.N, and Tb.Th and decreased Tb.Sp in the si-NRIR-SN group compared to the si-NC-SN group ( Figure 7C ). Additionally, histological analyses showed similar inflammatory infiltration and bone resorption in the LPS and si-NC-SN groups, whereas the si-NRIR-SN group had reduced inflammatory cytokine expression (IL-1β, IL-6, TNF-α) and elevated osteogenic marker expression (RUNX2, OPN, OCN) ( Figures 7D, E ). Together, NRIR knockdown modifies supernatant composition, reducing peri-implant inflammation and bone loss.