CX3CL1 promotes M1 macrophage polarization and osteoclast differentiation via NSUN5-mediated m5C modification

Abstract

Bone homeostasis refers to a dynamic equilibrium maintained between osteogenesis and osteoclastic bone resorption within the skeletal system. CX3CL1 (Fractalkine) is a chemokine that plays a significant regulatory role in bone homeostasis. This study aimed to investigate the mechanisms by which CX3CL1 regulates bone homeostasis. The expression of CX3CL1 in LPS-stimulated and RANKL-stimulated macrophages was examined using qPCR and Western blotting. Functional studies employed shRNA-mediated knockdown and overexpression of CX3CL1/NSUN5, followed by analysis of pro-inflammatory factor levels(IL-1β, IL-6, iNOS, and TNF-α), M1/M2 markers (CD86/CD206), osteoclast activity (TRAP staining, CTX-1 level), and key osteoclastogenic factors (NFATc1, c-Fos). Potential mechanisms were validated using Methylated RNA Immunoprecipitation (MeRIP), RNA Immunoprecipitation (RIP), and Dual-Luciferase Reporter Assay experiments. An ovariectomy (OVX)-induced osteoporosis mouse model was used for in vivo validation. Results showed that CX3CL1 was significantly upregulated in LPS- and RANKL-stimulated RAW 264.7 cells. Knockdown of CX3CL1 inhibited macrophage M1 polarization and osteoclast differentiation. NSUN5 interacted with CX3CL1 and suppressed its stability by promoting the m5C modification of CX3CL1 mRNA. Additionally, Overexpression of CX3CL1 reversed the inhibitory effect of NSUN5 overexpression on macrophage M1 polarization and osteoclast differentiation. In OVX mice, NSUN5 overexpression preserved bone mass (increased BV/TV, reduced Tb.Sp), while CX3CL1 co-expression abolished this protection. In conclusion, CX3CL1 accelerates M1 macrophage polarization and promotes osteoclast differentiation, mechanistically regulated by m5C modification mediated by NSUN5. This study provides novel therapeutic strategies and targets for maintaining bone homeostasis and preventing and treating bone-related diseases.

Introduction

Bone homeostasis refers to the capability of the skeletal system to maintain its structural and functional stability throughout growth, development, maturation, and aging. This process involves intricate cellular activities and molecular regulatory mechanisms, primarily the balance between osteogenesis by osteoblasts and osteoclastic bone resorption by osteoclasts¹. When bone homeostasis is disrupted, it can lead to metabolic bone disorders, such as osteoporosis, a prevalent condition characterized by decreased bone mass and microarchitectural deterioration, resulting in increased fracture risk². Macrophages, as key regulators of the bone microenvironment, play a pivotal role in bone remodeling by modulating inflammation and influencing osteoclast differentiation³. Macrophage polarization—particularly the pro-inflammatory M1 phenotype—exacerbates osteoclast activation, thereby contributing to pathological bone loss⁴. However, the precise molecular mechanisms linking macrophage polarization to osteoclastogenesis remain incompletely understood.

Osteogenic differentiation is the process by which mesenchymal stem cells develop into functional osteoblasts capable of synthesizing and mineralizing bone matrix, primarily mediated by the Wnt/β-catenin, BMP, and Notch signaling pathways. This process serves as the fundamental mechanism of bone formation^5,6. In contrast, osteoclasts, which are multinucleated cells originating from monocyte/macrophage precursors, mediate bone degradation through RANKL and M-CSF signaling, with osteoprotegerin (OPG) serving as a critical negative regulator of osteoclast overactivation^7,8. Pathological conditions including postmenopausal osteoporosis often involve disrupted communication between osteoblasts and osteoclasts, where inflammatory cytokines promote excessive bone resorption and progressive bone loss⁹. While the fundamental roles of osteoblasts and osteoclasts in bone remodeling are well established, the key regulators governing osteogenic differentiation and osteoclast activation require further investigation.

Fractalkine/CX3C chemokine ligand 1 (CX3CL1) is a unique chemokine that is expressed in various tissues and cell types, including vascular endothelial cells, smooth muscle cells, macrophages, and osteoblasts¹⁰. CX3CL1 exists not only on the cell membrane but can also be cleaved into a soluble form and released into the extracellular environment through the action of metalloproteinases. CX3CL1 mediates a variety of cellular functions, including cell migration, adhesion, and signaling, by binding to its unique receptor, CX3CR1¹¹. In recent years, significant advancements have been made in the research of CX3CL1 regarding its roles in immune regulation¹², neuroprotection¹³, and cardiovascular diseases¹⁴. Notably, CX3CL1 enhances osteoclast differentiation and bone resorption¹⁵, yet its regulatory mechanisms—particularly at the post-transcriptional level—remain elusive. Intriguingly, RNA methylation (m5C), catalyzed by NOP2/Sun RNA methyltransferase family members (NSUNs), has emerged as a critical epigenetic regulator of gene expression in various biological processes¹⁶. Whether m5C modification influences CX3CL1 expression and, consequently, macrophage-osteoclast communication is unknown.

In our study, we investigated the expression of CX3CL1 in LPS-stimulated and RANKL-stimulated macrophages, uncovered the promoting effects of CX3CL1 on the M1 macrophages and activation of osteoclasts, and revealed that the underlying mechanism is associated with the regulatory role of the NSUN5 factor. This study provides a promising regulator for the balance of bone homeostasis and provides a new perspective for the prevention and treatment of bone-related diseases.

Methods

Cell culture and treatment

The murine macrophage cell line RAW 264.7 (ATCC, Manassas, VA, USA) was cultured in complete DMEM medium (Gibco, Waltham, MA, USA) containing 10% FBS and 1% penicillin-streptomycin at 37 °C with 5% CO₂. For experiments, cells were plated in 6-well plates at 5 × 10⁵ cells/well and allowed to adhere for 24 h. M1 polarization was stimulated by 24 h treatment with 100 ng/mL LPS (Sigma-Aldrich, St. Louis, MO, USA). Osteoclast differentiation was achieved through 5 day culture with 50 ng/mL RANKL (R&D Systems, Minneapolis, MN, USA) and 25 ng/mL M-CSF (PeproTech, Rocky Hill, NJ, USA).

Cell transfection

Short hairpin (sh) RNA targeting CX3CL1 (shCX3CL1), shRNA negative control (sh-NC), shRNA targeting NSUN2-7, NSUN5 overexpressing plasmids, CX3CL1 overexpressing plasmids, and empty vector were obtained from Shanghai GenePharma company. These plasmids were transfected into RAW264.7 cells using Lipofectamine 3000 (Invitrogen, Carlsbad, CA, USA).

shCX3CL1: sense: 5′-AGGAGATAAACCCAGTTCATA-3′; anti-sense: 5′-TATGAACTGGGTTTATCTCCT-3′.

shCX3CR1: sense: 5′-CCCTTGCTTATCATGAGCTTT-3′; anti-sense: 5′-AAAGCTCATGATAAGCAAGGG-3′.

shNSUN2: sense: 5′-CCTGAAGATGATCCTTTATTT-3′; anti-sense: 5′-AAATAAAGGATCATCTTCAGG-3′.

shNSUN3: sense: 5′-CGGCAATAACTGAACTATATT-3′; anti-sense: 5′-AATATAGTTCAGTTATTGCCG-3′.

shNSUN4: sense: 5′-GCTGGTAATACCAAACCTCAT-3′; anti-sense: 5′-ATGAGGTTTGGTATTACCAGC-3′.

shNSUN5: sense: 5′-CTGCCTGAGCTTCTTGTATTT-3′; anti-sense: 5′-AAATACAAGAAGCTCAGGCAG-3′.

shNSUN6: sense: 5′-CGATGCAACAAAGGCACTTAA-3′; anti-sense: 5′-TTAAGTGCCTTTGTTGCATCG-3′.

shNSUN7: sense: 5′-CACTCTGTGAAGGCTTTGATA-3′; anti-sense: 5′-TATCAAAGCCTTCACAGAGTG-3′.

shNC: sense: 5′-CAACAAGATGAAGAGCACCAA-3′; anti-sense: 5′-TTGGTGCTCTTCATCTTGTTG-3′.

Quantitative real-time PCR (qPCR)

Gene-specific primers were designed through Primer-BLAST (NCBI) and commercially obtained from IDT (Coralville, IA). Total RNA was isolated from treated RAW264.7 cells using RNeasy Mini Kit (Qiagen, CA), followed by cDNA synthesis from 1 µg RNA with a reverse transcription kit (Thermo Fisher, MA). Quantitative PCR analysis was performed on a Bio-Rad CFX96 system, with gene expression quantified using the 2^−∆∆CT method.

CX3CL1: Forward Primer: 5’-CTGGCCGCGTTCTTCCATT-3’ Reverse Primer: 5’-GCACATGATTTCGCATTTCGT-3’.

IL-1β: Forward Primer: 5’-GAAATGCCACCTTTTGACAGTG-3’ Reverse Primer: 5’-TGGATGCTCTCATCAGGACAG-3’.

IL-6: Forward Primer: 5’-CTGCAAGAGACTTCCATCCAG-3’ Reverse Primer: 5’-AGTGGTATAGACAGGTCTGTTGG-3’.

iNOS: Forward Primer: 5’-GTTCTCAGCCCAACAATACAAGA-3’ Reverse Primer: 5’-GTGGACGGGTCGATGTCAC-3’.

TNF-α: Forward Primer: 5’-CAGGCGGTGCCTATGTCTC-3’ Reverse Primer: 5’-CGATCACCCCGAAGTTCAGTAG-3’.

NSUN2: Forward Primer: 5’-AGGTGGCTATCCCGAGATCG-3’ Reverse Primer: 5’-GACTCCATGAATTGGTCCCATT-3’.

NSUN3: Forward Primer: 5’-CAATATGCCATCCTCTTCAACCG-3’ Reverse Primer: 5’-AGGACTGTGTGATAGCCCCTC-3’.

NSUN4: Forward Primer: 5’-TGGGATAGTGTGAGTGCTAAGC-3’ Reverse Primer: 5’-AAGCATCGAAGATTTGGGCTG-3’.

NSUN5: Forward Primer: 5’-ACCTGAAGCAGTTGTACGCTC-3’ Reverse Primer: 5’-CCCCTTCCCCAGCAATAATTC-3’.

NSUN6: Forward Primer: 5’-AAGACAACAGGGTGAAGTGATTG-3’ Reverse Primer: 5’-TCCATCAAATTCTTTGGCTCCTT-3’.

NSUN7: Forward Primer: 5’-TCTCAAGGTGGTCTACCGAAA-3’ Reverse Primer: 5’-TTCATTGCGTGTGTTAGCTGT-3’.

GAPDH: Forward Primer: 5’-AGGTCGGTGTGAACGGATTTG-3’ Reverse Primer: 5’-GGGGTCGTTGATGGCAACA-3’.

Flow cytometry analysis and macrophage subset quantification

Single-cell suspensions were prepared by enzymatic digestion of tissues using collagenase I (1 mg/ml) and dispase II (1 mg/ml) at 37 °C for 30 min, followed by red blood cell lysis and filtration through a 70-µm strainer. Cells were stained with anti-mouse antibodies against CD45 (clone 30-F11), CD11b (clone M1/70), CD86 (clone GL-1), and CD206 (clone C068C2) (all from BioLegend, San Diego, CA, USA) in 100 µl PBS for 30 min on ice, then fixed and permeabilized using the Foxp3/Transcription Factor Staining Buffer Set. Flow cytometry was performed on a BD LSRFortessa (BD Biosciences, San Jose, CA, USA), and macrophage subsets were identified as CD45 + CD11b + CD86+ (M1-like) and CD45 + CD11b + CD206+ (M2-like) populations.

Enzyme-linked immunosorbent assay

C-terminal telopeptide of type I collagen (CTX-1) levels in cell culture supernatants were measured using a commercial mouse CTX-1 ELISA kit (Cusabio, Wuhan, Hubei, China; catalog #CSB-E12787m). Briefly, 100 µL of undiluted supernatant samples and standards were added to pre-coated wells and incubated for 2 h at 37 °C. After washing with PBS containing 0.05% Tween-20, biotinylated detection antibody was added, followed by streptavidin-HRP incubation. The reaction was developed with TMB substrate, stopped with 2 N H₂SO₄, and absorbance was measured at 450 nm using a SpectraMax M5 microplate reader (Molecular Devices, San Jose, CA, USA). CTX-1 concentrations were determined from the standard curve (0.156-10 ng/mL) using four-parameter logistic regression.

Western blot

The proteins from the RAW264.7 cells were extracted using conventional methods. Protein concentration was determined using the BSA kit. The protein was separated by SDS-PAGE gels and then transferred to the PVDF membrane. Next, the protein rapid-blocking solution was added to block the membranes at 37 °C for 20 min. The membranes were incubated overnight at 4 °C with primary antibodies at 1: 1000 dilution and then incubated with secondary antibodies (Abcam, Cambridge, MA, USA) for 2 h. Bands were imaged with a ChemiDoc MP Imaging System (Bio-Rad Laboratories). The information of primary antibodies is as follows: anti-CX3CL1 (ab25088), anti-IL-1β (ab283818), anti-IL-6 (ab290735), anti-iNOS (ab178945), anti-TNF-α (ab183218), anti-NSUN5 (34405, Cell Signaling Technology, Danvers, MA, USA), anti-p-IKKαβ (ab194528), anti-IKKαβ (ab32041), anti-p-p65 (ab76302), anti-p65 (ab32536), anti-NFATc1 (5861, Cell Signaling Technology), anti-TRAP (ab133238), anti-c-Fos (ab208942), anti-APC5 (ab72516), and anti-GAPDH (ab8245).

TRAP staining

According to the experimental design, cells from different treatment groups were fixed with 4% paraformaldehyde for 15 min. Following the manufacturer’s instructions (TRAP staining kit; Sigma-Aldrich, St. Louis, MO, USA), the cells were incubated with 500 µL of TRAP staining solution at 37 °C for 1 h. After washing the cells with distilled water, they were counterstained with 500 µL of hematoxylin solution for 1–2 min to stain the nuclei. TRAP-positive multinucleated cells (osteoclasts) were identified under light microscopy and quantified using ImageJ software.

Methylated RNA Immunoprecipitation (MeRIP)

The m5C level of CX3CL1 was determined using a GenSeq^® m5C MeRIP kit (GS-ET-003, Cloudseq, Shanghai, China) according to the manufacturer’s protocols. The total RNA was extracted and fragmented. The magnetic beads were incubated with 2 µL m5C or IgG antibody for 1 h. Nuclease-free water (200 µL) containing RNA fragments was added to the prepared magnetic beads to incubate for 1 h for immunoprecipitation. After washing with IP buffer, the RNA was purified, and gene expression was detected by qPCR.

RIP assays

EZ-Magna RIP RNA-Binding Protein Immunoprecipitation Kit from Shanghai Millipore was used to verify the interaction between NSUN5 and CX3CL1. Briefly, the cell lysates lysed with RIP lysis buffer were incubated with magnetic beads conjugated with anti-NSUN5 (ab121633) or anti-IgG (ab133470) antibodies at 4℃ overnight. After purification, RNA was extracted from the mixture and analyzed by qPCR.

Dual-luciferase reporter assay

The pGL4 luciferase reporter vector (Promega, Madison, WI, USA) was constructed with either wild-type (WT) or mutant (MUT) CX3CL1 sequences at three specific sites. RAW264.7 cells were co-transfected with WT/MUT reporter plasmids, ShNC/shNSUN5, and pRL-TK vector (Promega) using Lipofectamine 3000 (Invitrogen). Luciferase activity was measured using the Dual-Luciferase reporter system (Promega), with firefly luciferase signals normalized to Renilla luminescence.

RNA stability detection

The qPCR assay was used to analyze the CX3CL1 mRNA stability after RAW264.7 cells were treated with 5 µg/mL actinomycin D (Merck, Darmstadt, Germany) for 0, 6, 12, and 24 h.

Osteoporosis mouse model

Female C57BL/6J mice (10-week-old, n = 6/group) were purchased from Charles River Laboratories (Wilmington, MA, USA) and acclimatized for 1 week under specific pathogen-free conditions (24 ± 1 °C, 55 ± 5% humidity, 12 h light/dark cycle). The mice were randomly divided into six groups (n = 6 per group): (1) Sham (sham-operated), (2) ovariectomy (OVX), (3) OVX + Vector, (4) OVX + NSUN5, (5) OVX + NSUN5 + Vector, and (6) OVX + NSUN5 + CX3CL1. The lentiviral vectors mediating NSUN5 and CX3CL1 overexpression were synthesized by GenePharma with a titer of 1 × 10⁸ TU/mL. Viral particles were delivered via tail vein injection twice weekly for 8 weeks. Mice were anesthetized with 2% isoflurane (Sigma-Aldrich) and subjected to bilateral OVX through dorsal approach under aseptic conditions, while sham-operated controls underwent identical procedures without ovary removal; postoperatively, mice received buprenorphine analgesia (0.1 mg/kg, Reckitt Benckiser, Slough, UK) and were monitored for 72 h. At 12 weeks post-surgery, mice were euthanized with 5% isoflurane, femurs were harvested and scanned using micro-computed tomography (µCT, Skyscan 1176, Bruker, Belgium) to evaluate bone microarchitecture. Three-dimensional reconstruction and analysis were performed to quantify bone mineral density (BMD, g/cm³), trabecular separation/spacing (Tb. Sp, mm), and bone volume/tissue volume ratio (BV/TV, %). This study was approved by the Ethics Committee of Songgang People’s Hospital. All animal experiments were complied with the ARRIVE guidelines. All methods were carried out in accordance with relevant guidelines and regulations.

Statistical analysis

Statistical analyses were performed using GraphPad Prism 7 (La Jolla, CA) and SPSS 13.0 (Chicago, IL). Data from triplicate independent experiments were expressed as mean ± SD and analyzed by Student’s t-test, one-way ANOVA, or one-way ANOVA, with statistical significance set at p < 0.05.

Results

CX3CL1 is highly expressed in vitro

To investigate the expression pattern of CX3CL1 in RAW 264.7 cells under LPS and RANKL stimulation, both qPCR and western blot analysis were conducted. The experimental data demonstrated a marked increase in both transcriptional activity and protein abundance of CX3CL1 in stimulated RAW 264.7 cells (Fig. 1A-D).

[See PDF for image]

Fig. 1

CX3CL1 was highly expressed in vitro. (A-B) CX3CL1 expression was assessed in LPS-stimulated RAW 264.7 cells after 24-h treatment using qPCR for mRNA levels and Western blot analysis for protein expression. (C-D) CX3CL1 expression was assessed in RANKL-stimulated RAW 264.7 cells after 5-days treatment using qPCR for mRNA levels and Western blot analysis for protein expression. Data were analyzed by Student’s t-test. All data are presented as the mean ± SD. ^**p < 0.01. (n = 3).

Knockdown of CX3CL1 inhibits macrophage M1 polarization and osteoclast differentiation

Next, we explored the function of CX3CL1 in LPS-stimulated RAW 264.7 cells. The transfection efficiency was detected using qPCR after the transfection of shCX3CL1 into the RAW 264.7 cells (Fig. 2A). CX3CL1 knockdown in LPS-stimulated RAW 264.7 macrophages resulted in significant reduction of pro-inflammatory mediators, including IL-1β, IL-6, iNOS, and TNF-α (Fig. 2B-F). We then examined the expression of CD86 (an M1 marker) and CD206 (an M2 marker) using flow cytometry (Fig. 2G). The results demonstrated that silencing CX3CL1 reduced the proportion of CD86⁺ cells while increasing the proportion of CD206⁺ cells, indicating that CX3CL1 depletion attenuates M1 phenotype activation in macrophages. To investigate the regulatory role of CX3CL1 in osteoclast function, we measured the levels of the bone resorption marker CTX-1 by ELISA. The results showed that LPS-stimulated RAW 264.7 cells exhibited increased CTX-1 levels, which were reversed by CX3CL1 knockdown (Fig. 2H). As is shown in Fig. 2I, TRAP staining revealed that RANKL remarkably increased multiple nuclear cells, indicating an enhanced capacity for mature osteoclast formation. After CX3CL1 silencing, the number of multiple nuclear cells was prominently reduced, suggesting that the down-regulation of CX3CL1 could inhibit the formation of osteoclasts stimulated by RANKL. Moreover, CX3CR1 receptor silencing significantly reduced the number of TRAP-positive multinucleated cells (Supplementary Fig. 1 A). Simultaneously, the expression of NFATc1, TRAP, c-Fos, and APC5, key mediators of osteoclast production, were analyzed by western blot. They all were remarkably upregulated in RANKL-stimulated RAW 264.7 cells. However, they are dramatically downregulated with the knockdown of the CX3CL1 (Fig. 2J). These results demonstrated that CX3CL1 silencing significantly suppressed osteoclast function. Our supplementary data demonstrate that CX3CL1 knockout suppresses NF-κB signaling activation, as evidenced by reduced protein expression of phosphorylated IKKα/β and p65 (Supplementary Fig. 1B).

[See PDF for image]

Fig. 2

Knockdown of CX3CL1 inhibited macrophage M1 polarization and osteoclast differentiation. (A) The transfection efficiency was measured by qPCR. (B-F) The mRNA levels and protein expression of IL-1β, IL-6, iNOS, and TNF-α in LPS-stimulated RAW 264.7 cells were detected using qPCR and western blot after CX3CL1 knockdown, respectively. (G) Flow cytometry analysis of CD86⁺ and CD206⁺ macrophage proportion in each group. (H) The level of CTX-1 was detected by ELISA assay. (I) The ability of mature osteoclast formation was analyzed by TRAP staining. (J) The expressions of NFATc1, TRAP, c-Fos, and APC5 in RANKL-stimulated RAW 264.7 cells were analyzed by western blot after CX3CL1 knockdown. Data were analyzed by Student’s t-test (A) or one-way ANOVA followed by Tukey’s post hoc test (B-J). All data are presented as the mean ± SD. ^**p < 0.01. (n = 3).

NSUN5 negatively regulates CX3CL1 expression in m5C modification manner

RNA modification not only participates in bone development but also plays an important role in maintaining bone homeostasis balance. The NSUN family of RNA methyltransferases (NSUN1-7) is a key enzyme in m5C modification. Herein, we knocked down NSUN2-7 and found that CX3CL1 mRNA level was significantly increased only when NSUN5 was silenced (Fig. 3A and B). The protein expression of NSUN5 was decreased in LPS-stimulated RAW 264.7 cells (Fig. 3C). Next, the m5C modification level of CX3CL1 mRNA was decreased distinctly after NSUN5 knockdown (Fig. 3D). The interaction between CX3CL1 and NSUN5 was verified by RIP assay (Fig. 3E). The m5C site of CX3CL1 mRNA was predicted using the website RNA m5C finder (Fig. 3F). We selected the three most likely sites (538th, 1286th, and 1329th) for mutation to perform luciferase experiments. The results indicated that silencing NSUN5 only drastically weakened the luciferase activity of CX3CL1-wt containing 1286 and 1329 sites. In comparison, the luciferase activity of CX3CL1-mut did not change obviously in RAW 264.7 cells (Fig. 3G). Moreover, the silencing of NSUN5 enhanced the CX3CL1 mRNA stability (Fig. 3H). These findings confirmed that NSUN5 inhibited the CX3CL1 mRNA expression by promoting the m5C modification of CX3CL1, indicating the negative relationship between CX3CL1 and NSUN5.

[See PDF for image]

Fig. 3

NSUN5 negatively regulated CX3CL1 expression in m5C modification manner. (A) The transfection efficiency was measured by qPCR. (B) The levels of CX3CL1 mRNA were measured using qPCR. (C) The protein expression of NSUN5 in RANKL-stimulated RAW 264.7 cells were measured using Western blot. (D) The m5C modification level of CX3CL1 was detected by MeRIP. (E) The interaction between NSUN5 and CX3CL1 was measured by RIP. (F) The prediction of m5C modification sites in CX3CL1. (G) The binding relationship between NSUN5 and CX3CL1 was analyzed by Luciferase assay. (H) The stability of CX3CL1 mRNA was measured by qPCR after RAW 264.7 cells were treated with actinomycin D at 0, 6, 12, and 24 h. Data were analyzed by Student’s t-test (A, B, D, G), one-way ANOVA (E) or two-way ANOVA (H). All data are presented as the mean ± SD. ^*p < 0.05 and ^**p < 0.01. (n = 3).

Overexpression of CX3CL1 reverses the inhibitory effect of NSUN5 overexpression on the macrophage M1 polarization and osteoclast differentiation in RAW 264.7 cells

Finally, we demonstrated the role of CX3CL1 and NSUN5 in macrophage behavior through rescued experiments. The transfection efficiency was measured by qPCR after the transfection of NSUN5 and CX3CL1 overexpression plasmid into the RAW 264.7 cells, respectively (Fig. 4A and B). The mRNA levels and protein expression of IL-1β, IL-6, iNOS, and TNF-α were decreased obviously after the overexpression of NSUN5 in LPS-stimulated RAW 264.7 cells and therewith recovered by CX3CL1 overexpression (Fig. 4C-G). Overexpression of NSUN5 increased the proportion of CD86⁺ cells and decreased the proportion of CD206⁺ cells, an effect that was reversed by CX3CL1 overexpression (Fig. 4H). Similarly, the reduced CTX-1 levels induced by NSUN5 overexpression were restored upon CX3CL1 overexpression (Fig. 4I). In addition, the capacity for mature osteoclast formation weakened by overexpression of NSUN5 was recovered after CX3CL1 overexpression (Fig. 4J). Overexpression of CX3CL1 abolished the inhibitory influences of NSUN5 overexpression on the expression of NFATc1, TRAP, c-Fos, and APC5 in RANKL-stimulated RAW 264.7 cells (Fig. 4K). Based on these data, we concluded that NSUN5 inhibited macrophage M1 polarization and osteoclast differentiation by regulating the expression of CX3CL1.

[See PDF for image]

Fig. 4

Overexpression of CX3CL1 reversed the inhibitory effect of NSUN5 overexpression on the macrophage M1 polarization and osteoclast differentiation in RAW 264.7 cells. (A-B) The transfection efficiency was analyzed using qPCR. (C-G) The mRNA levels of IL-1β, IL-6, iNOS, and TNF-α in LPS-stimulated RAW 264.7 cells were detected using qPCR. (H) Flow cytometry analysis of CD86⁺ and CD206⁺ macrophage proportion in each group. (I) The level of CTX-1 was detected by ELISA assay. (J) The ability of mature osteoclast formation was analyzed by TRAP staining. (K) The expressions of NFATc1, TRAP, c-Fos, and APC5 in RANKL-stimulated RAW 264.7 cells were analyzed by western blot. Data were analyzed by Student’s t-test (A-B) or one-way ANOVA followed by Tukey’s post hoc test (C-K). All data are presented as the mean ± SD. ^**p < 0.01. (n = 3).

CX3CL1 abrogates the bone-protective effects of NSUN5 in OVX-induced osteoporotic mice

To investigate the in vivo roles of CX3CL1 and NSUN5 in osteoporosis pathogenesis, we established a postmenopausal osteoporosis model through bilateral OVX in C57BL/6 mice. Representative micro-CT reconstructions (Fig. 5A) demonstrated severe trabecular bone loss with disconnected trabecular networks in OVX controls, while NSUN5-overexpressing mice exhibited preserved trabecular architecture. Quantitative analysis demonstrated that compared to the OVX group, NSUN5 overexpression significantly attenuated OVX-induced osteoporosis, as evidenced by increased bone mineral density (BMD) and bone volume fraction (BV/TV). This protective effect was effectively reversed by CX3CL1 co-overexpression (Fig. 5B and C). Furthermore, NSUN5 overexpression markedly reduced trabecular separation (Tb.Sp), an improvement that was completely abolished by concurrent CX3CL1 overexpression (Fig. 5D).

[See PDF for image]

Fig. 5

CX3CL1 abrogated the bone-protective effects of NSUN5 in OVX-induced osteoporotic mice. (A-B) The effects of CX3CL1 and NSUN5 on (A) Micro-CT, (B) bone mineral density (BMD), (C) bone volume fraction (BV/TV), and (D) trabecular separation (Tb.Sp) of femurs. Data were analyzed by one-way ANOVA followed by Tukey’s post hoc test (B-D). All data are presented as the mean ± SD. ^**p < 0.01. (n = 6).

Discussion

Recent studies have demonstrated that chemokines play a crucial role in the regulation of bone homeostasis¹⁷. Chemokines are small protein molecules that attract and guide the migration of immune cells, playing important roles in various physiological and pathological processes, including immune responses, inflammation, and tissue repair¹⁸. Among them, CX3CL1 is the chemokine most closely associated with bone homeostasis. LU et al. demonstrated that loss of CX3CL1 accelerates osteogenic differentiation¹⁹. CX3CL1 is involved in the migration of osteoclast precursors and osteoclast formation²⁰. For instance, CX3CL1 induces chemotaxis and cell adhesion of osteoclast precursors²¹. The osteoclastic differentiation capacity is regulated by RANKL and M-CSF. Here, using RANKL-stimulated macrophages²², we observed a significant increase in CX3CL1 levels. Knockdown of CX3CL1 expression not only inhibited the formation of mature osteoclasts but also downregulated the expression of key mediators of osteoclastogenesis, including CTX-1, NFATc1, TRAP, c-Fos, and APC5. Supplementary data further demonstrated that knockdown of CX3CR1 significantly reduced the number of TRAP-positive multinucleated cells, suggesting that inhibition of the CX3CL1-CX3CR1 axis suppresses osteoclast differentiation.

Macrophages are a critical component of the immune system, characterized by high plasticity and functional diversity. M1-type macrophages are typically associated with pro-inflammatory responses and host defense, whereas M2-type macrophages are linked to anti-inflammatory responses and tissue repair²³. Moreover, the regulation of the immune system plays a significant role in bone homeostasis, particularly the polarization of M1-type macrophages^24,25. M1-type macrophages are primarily activated by stimuli such as LPS and IFN-γ. M1 macrophages produce large amounts of pro-inflammatory cytokines (such as TNF-α, IL-1β, and IL-6) and ROS, participating in host defense and inflammatory responses²⁶. Studies have reported that high levels of pro-inflammatory factors inhibit osteoblast differentiation and function, reduce bone matrix synthesis, and promote osteoclast differentiation and activation, increasing their resorption of bone matrix²⁷. For example, IL-1β promotes osteoclast formation by increasing the expression of IGF2 and CX3CL1 in non-osteoclasts²⁸. When the polarization of M1 macrophages is imbalanced, it can lead to the occurrence of various diseases related to bone homeostasis imbalance. CX3CL1 enhances LPS-stimulated macrophage activation via the Wnt/β-catenin signaling pathway²⁹. Liu et al. determined that high expression of CX3CL1 in macrophages promotes M1 polarization³⁰. Notably, Feng et al. showed that up-regulated CX3CL1 promotes M1-type macrophage polarization and osteoclast differentiation through the NF-κB signaling pathway³¹. In addition, inhibition of the CX3CL1-CX3CR1 pathway is a potential therapeutic strategy for inflammatory/immune diseases and bone destruction³². This study used LPS-stimulated macrophages to mimic the M1 polarization process, during which we found that CX3CL1 was highly expressed. Concurrently, silencing the expression of CX3CL1 inhibited the levels of IL-1β, IL-6, iNOS, and TNF-α, indicating that CX3CL1 promotes M1 cell polarization. Furthermore, our supplementary data demonstrated that suppression of CX3CL1 expression inhibited NF-κB signaling activation, suggesting that the biological functions of CX3CL1 are likely mediated through targeting the NF-κB pathway.

m5C modification is an important type of RNA modification that alters the structure and function of RNA, participating in various biological processes³³. An imbalance in m5C modification can affect the function of osteoblasts and osteoclasts, leading to increased bone destruction³⁴. The NSUN family is a group of RNA methyltransferases primarily responsible for catalyzing m5C modifications on RNA molecules³⁵. The NSUN family includes multiple members, NSUN1-7, among which NSUN2 plays crucial roles in bone homeostasis³⁴. In this study, to explore the specific potential mechanisms of CX3CL1, we found that only the knockdown of NSUN5 expression significantly increased the mRNA level of CX3CL1, while the level of m5C modification on CX3CL1 was markedly reduced. Moreover, CX3CL1 interacted with NSUN5, and the silencing of NSUN5 enhanced the stability of CX3CL1 mRNA. These findings confirm that NSUN5 negatively regulates the mRNA expression of CX3CL1 by promoting its m5C modification. In osteoclastogenesis assays, we found that overexpression of NSUN5 attenuates the ability to form mature osteoclasts. However, overexpression of CX3CL1 reverses the inhibitory effect of NSUN5 overexpression on the formation of mature osteoclasts. Based on the role of M1 polarization in bone homeostasis, as expected, after overexpression of NSUN5, M1 macrophage polarization was inhibited, and overexpression of CX3CL1 abolished this phenomenon.

The clinical relevance of our findings is underscored by the OVX mouse model results. NSUN5 overexpression significantly ameliorated osteoporosis symptoms, while CX3CL1 co-expression abolished these protective effects. This suggests that pharmacological modulation of the NSUN5-CX3CL1 axis could represent a promising therapeutic strategy for osteoporosis, particularly in postmenopausal women where excessive osteoclast activity drives bone loss.

Despite these findings, this study has several limitations. First, it is unclear whether NSUN5 regulates CX3CL1 mRNA stability through classical decay pathways (e.g., YTHDF2-mediated decay) or via altered ribonuclease sensitivity. Further studies should identify the specific RNA-binding proteins that recognize m5C-modified CX3CL1 mRNA and their downstream decay machinery. Second, the cellular source of bioactive CX3CL1 and whether it functions through autocrine or paracrine mechanisms. Additionally, while the OVX mouse model provided evidence for the roles of CX3CL1 and NSUN5 in osteoporosis, the translational relevance to human disease remains to be validated. Finally, this study primarily relied on RAW 264.7 cells, and the findings were not validated in primary cells, particularly in LPS/RANKL-stimulated bone marrow-derived macrophages (BMMs), which may better reflect physiological conditions.

Future research should explore whether the NSUN5-CX3CL1 axis operates in osteoblasts or osteocytes, potentially revealing broader roles in bone remodeling. Further mechanistic work is needed to elucidate how m5C modification fine-tunes CX3CL1 mRNA expression and whether other RNA modifications contribute to bone metabolism. Alternative pathways regulating osteoclastogenesis, such as non-coding RNAs or additional post-transcriptional modifications, should also be investigated. Importantly, validation in primary cells is essential to confirm the translational relevance of our findings. These studies could pave the way for therapeutic strategies targeting RNA methylation in bone disorders.

In summary, this experiment elucidates from the perspective of the immune system the mechanism by which CX3CL1 promotes M1 macrophage polarization and, consequently, osteoclast differentiation. Our work establishes NSUN5-mediated m5C modification as a critical epigenetic regulator of CX3CL1 expression in bone homeostasis. These findings not only advance our understanding of osteoimmunology but also identify potential therapeutic targets for metabolic bone disorders.

Acknowledgements

Not applicable.

Author contributions

All authors participated in the design, interpretation of the studies and analysis of the data and review of the manuscript. Z D drafted the work and revised it critically for important intellectual content and was responsible for the acquisition, analysis and interpretation of data for the work; W Z made substantial contributions to the conception or design of the work. All authors read and approved the final manuscript.

Funding

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was approved by the Ethics Committee of Songgang People’s Hospital. All animal experiments were complied with the ARRIVE guidelines. All methods were carried out in accordance with relevant guidelines and regulations.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

1. Choi, JY. Healthy bone tissue homeostasis. Exp. Mol. Med.; 2020; 52, 1165.1:CAS:528:DC%2BB3cXhsF2ntLjP [DOI: https://dx.doi.org/10.1038/s12276-020-0472-3] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32788654][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8080694]

2. Chandra, A. & Rajawat, J. Skeletal aging and osteoporosis: mechanisms and therapeutics. Int. J. Mol. Sci.22https://doi.org/10.3390/ijms22073553 (2021).

3. Kong, L; Wang, Y; Smith, W; Hao, D. Macrophages in bone homeostasis. Curr. Stem Cell. Res. Ther.; 2019; 14, pp. 474-481.1:CAS:528:DC%2BC1MXhs1ensbvI [DOI: https://dx.doi.org/10.2174/1574888X14666190214163815] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30767753]

4. Yan, Q; Liu, H; Zhu, R; Zhang, Z. Contribution of macrophage polarization in bone metabolism: A literature review. Cytokine; 2024; 184, 156768.1:CAS:528:DC%2BB2cXitVykurzI [DOI: https://dx.doi.org/10.1016/j.cyto.2024.156768] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/39340960]

5. Ju, C et al. Mesenchymal stem cell-associated LncRNA in osteogenic differentiation. Biomed. Pharmacother; 2019; 115, 108912.1:CAS:528:DC%2BC1MXptFeltL8%3D [DOI: https://dx.doi.org/10.1016/j.biopha.2019.108912] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31048188]

6. Ahmadi, A et al. Recent advances on small molecules in osteogenic differentiation of stem cells and the underlying signaling pathways. Stem Cell. Res. Ther.; 2022; 13, 518.1:CAS:528:DC%2BB38XivVyqur7P [DOI: https://dx.doi.org/10.1186/s13287-022-03204-4] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36371202][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9652959]

7. Kaifu, T et al. DCIR suppresses osteoclastic proliferation and resorption by downregulating M-CSF and RANKL signaling. Front. Immunol.; 2023; 14, 1159058.1:CAS:528:DC%2BB3sXht1elu7nF [DOI: https://dx.doi.org/10.3389/fimmu.2023.1159058] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37266426][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10230091]

8. Udagawa, N et al. Osteoclast differentiation by RANKL and OPG signaling pathways. J. Bone Min. Metab.; 2021; 39, pp. 19-26.1:CAS:528:DC%2BB3cXitFamsbvL [DOI: https://dx.doi.org/10.1007/s00774-020-01162-6]

9. Fischer, V; Haffner-Luntzer, M. Interaction between bone and immune cells: implications for postmenopausal osteoporosis. Semin Cell. Dev. Biol.; 2022; 123, pp. 14-21. [DOI: https://dx.doi.org/10.1016/j.semcdb.2021.05.014] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34024716]

10. Ostuni, MA et al. CX3CL1 homo-oligomerization drives cell-to-cell adherence. Sci. Rep.; 2020; 10, 9069.2020NatSR.10.9069O1:CAS:528:DC%2BB3cXhtFWmu7bK [DOI: https://dx.doi.org/10.1038/s41598-020-65988-w] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32494000][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7271195]

11. Helmke, A et al. CX3CL1-CX3CR1 interaction mediates macrophage-mesothelial cross talk and promotes peritoneal fibrosis. Kidney Int.; 2019; 95, pp. 1405-1417.1:CAS:528:DC%2BC1MXmsVWmur8%3D [DOI: https://dx.doi.org/10.1016/j.kint.2018.12.030] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30948201]

12. Nanki, T; Imai, T; Kawai, S. Fractalkine/CX3CL1 in rheumatoid arthritis. Mod. Rheumatol.; 2017; 27, pp. 392-397.1:CAS:528:DC%2BC28Xhtlamur7P [DOI: https://dx.doi.org/10.1080/14397595.2016.1213481] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/27484962]

13. Subbarayan, MS; Joly-Amado, A; Bickford, PC; Nash, KR. CX3CL1/CX3CR1 signaling targets for the treatment of neurodegenerative diseases. Pharmacol. Ther.; 2022; 231, 107989.1:CAS:528:DC%2BB3MXitVKnu7vE [DOI: https://dx.doi.org/10.1016/j.pharmthera.2021.107989] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34492237]

14. Apostolakis, S; Spandidos, D. Chemokines and atherosclerosis: focus on the CX3CL1/CX3CR1 pathway. Acta Pharmacol. Sin; 2013; 34, pp. 1251-1256.1:CAS:528:DC%2BC3sXhsFOhsr7N [DOI: https://dx.doi.org/10.1038/aps.2013.92] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/23974513][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4002164]

15. Hoshino, A et al. Roles of chemokine receptor CX3CR1 in maintaining murine bone homeostasis through the regulation of both osteoblasts and osteoclasts. J. Cell. Sci.; 2013; 126, pp. 1032-1045.1:CAS:528:DC%2BC3sXotFaiuro%3D [DOI: https://dx.doi.org/10.1242/jcs.113910] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/23264747]

16. Chen, X et al. RNA m5C modification: from physiology to pathology and its biological significance. Front. Immunol.; 2025; 16, 1599305.1:CAS:528:DC%2BB2MXht12qt77K [DOI: https://dx.doi.org/10.3389/fimmu.2025.1599305] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/40370440][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12075115]

17. Ito, H. Chemokines in mesenchymal stem cell therapy for bone repair: a novel concept of recruiting mesenchymal stem cells and the possible cell sources. Mod. Rheumatol.; 2011; 21, pp. 113-121.1:CAS:528:DC%2BC3MXktl2jsbo%3D [DOI: https://dx.doi.org/10.1007/s10165-010-0357-8] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/20830500]

18. Cambier, S; Gouwy, M; Proost, P. The chemokines CXCL8 and CXCL12: molecular and functional properties, role in disease and efforts towards Pharmacological intervention. Cell. Mol. Immunol.; 2023; 20, pp. 217-251.1:CAS:528:DC%2BB3sXitlyksLY%3D [DOI: https://dx.doi.org/10.1038/s41423-023-00974-6] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36725964][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9890491]

19. Lu, Z; Zhang, A; Dai, Y. CX3CL1 deficiency ameliorates inflammation, apoptosis and accelerates osteogenic differentiation, mineralization in LPS-treated MC3T3-E1 cells via its receptor CX3CR1. Ann. Anat.; 2023; 246, 152036. [DOI: https://dx.doi.org/10.1016/j.aanat.2022.152036] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36436718]

20. Matsuura, T et al. Involvement of CX3CL1 in the migration of osteoclast precursors across osteoblast layer stimulated by Interleukin-1ss. J. Cell. Physiol.; 2017; 232, pp. 1739-1745.1:CAS:528:DC%2BC28XhsVegtrrJ [DOI: https://dx.doi.org/10.1002/jcp.25577] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/27579490]

21. Koizumi, K et al. Role of CX3CL1/fractalkine in osteoclast differentiation and bone resorption. J. Immunol.; 2009; 183, pp. 7825-7831.1:CAS:528:DC%2BD1MXhsFCkt7vL [DOI: https://dx.doi.org/10.4049/jimmunol.0803627] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/19923448]

22. Song, C et al. Evaluation of efficacy on RANKL induced osteoclast from RAW264.7 cells. J. Cell. Physiol.; 2019; 234, pp. 11969-11975.1:CAS:528:DC%2BC1cXitl2hsb7F [DOI: https://dx.doi.org/10.1002/jcp.27852] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30515780]

23. Shapouri-Moghaddam, A et al. Macrophage plasticity, polarization, and function in health and disease. J. Cell. Physiol.; 2018; 233, pp. 6425-6440.1:CAS:528:DC%2BC1cXjsFWqs7s%3D [DOI: https://dx.doi.org/10.1002/jcp.26429] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/29319160]

24. Yao, Y et al. The Macrophage-Osteoclast Axis in osteoimmunity and Osteo-Related diseases. Front. Immunol.; 2021; 12, 664871.1:CAS:528:DC%2BB3MXpvVyhtL4%3D [DOI: https://dx.doi.org/10.3389/fimmu.2021.664871] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33868316][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8044404]

25. Hu, K; Shang, Z; Yang, X; Zhang, Y; Cao, L. Macrophage polarization and the regulation of bone immunity in bone homeostasis. J. Inflamm. Res.; 2023; 16, pp. 3563-3580.1:CAS:528:DC%2BB3sXhvVentb3M [DOI: https://dx.doi.org/10.2147/JIR.S423819] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37636272][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10460180]

26. Orecchioni, M; Ghosheh, Y; Pramod, AB; Ley, K. Macrophage polarization: different gene signatures in M1(LPS+) vs. Classically and M2(LPS-) vs. Alternatively activated macrophages. Front. Immunol.; 2019; 10, 1084.1:CAS:528:DC%2BB3cXhs1ahsrg%3D [DOI: https://dx.doi.org/10.3389/fimmu.2019.01084] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31178859][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6543837]

27. Mo, Q; Zhang, W; Zhu, A; Backman, LJ; Chen, J. Regulation of osteogenic differentiation by the pro-inflammatory cytokines IL-1beta and TNF-alpha: current conclusions and controversies. Hum. Cell; 2022; 35, pp. 957-971.1:CAS:528:DC%2BB38XhsV2gurbO [DOI: https://dx.doi.org/10.1007/s13577-022-00711-7] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35522425]

28. Otsuka, Y et al. IL-1beta promotes osteoclastogenesis by increasing the expression of IGF2 and chemokines in non-osteoclastic cells. J. Pharmacol. Sci.; 2023; 151, pp. 1-08.1:CAS:528:DC%2BB38XislCltL3E [DOI: https://dx.doi.org/10.1016/j.jphs.2022.10.007] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36522118]

29. Gong, Q; Jiang, Y; Pan, X; You, Y. Fractalkine aggravates LPS-induced macrophage activation and acute kidney injury via Wnt/beta-catenin signalling pathway. J. Cell. Mol. Med.; 2021; 25, pp. 6963-6975.1:CAS:528:DC%2BB3MXhsFaqu7bM [DOI: https://dx.doi.org/10.1111/jcmm.16707] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34101346][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8278080]

30. Liu, J et al. Mesenchymal stem cells elicits Anti-PD1 immunotherapy by targeted delivery of CX3CL1. Front. Pharmacol.; 2023; 14, 1136614.1:CAS:528:DC%2BB3sXkvV2ksrk%3D [DOI: https://dx.doi.org/10.3389/fphar.2023.1136614] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36843945][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9944415]

31. Feng, X et al. CX3CL1 promotes M1 macrophage polarization and osteoclast differentiation through NF-kappaB signaling pathway in ankylosing spondylitis in vitro. J. Transl Med.; 2023; 21, 573.1:CAS:528:DC%2BB3sXhvVCjsbvF [DOI: https://dx.doi.org/10.1186/s12967-023-04449-0] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37626378][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10463543]

32. Imai, T; Yasuda, N. Therapeutic intervention of inflammatory/immune diseases by Inhibition of the fractalkine (CX3CL1)-CX3CR1 pathway. Inflamm. Regen; 2016; 36, 9.1:CAS:528:DC%2BC1MXjvFehsbg%3D [DOI: https://dx.doi.org/10.1186/s41232-016-0017-2] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/29259682][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5725656]

33. Bohnsack, K. E., Hobartner, C. & Bohnsack, M. T. Eukaryotic 5-methylcytosine (m(5)C) RNA methyltransferases: mechanisms, cellular functions, and links to disease. Genes (Basel). 10https://doi.org/10.3390/genes10020102 (2019).

34. Yu, M et al. Aberrant NSUN2-mediated m5C modification of Exosomal LncRNA MALAT1 induced RANKL-mediated bone destruction in multiple myeloma. Commun. Biol.; 2024; 7, 1249.1:CAS:528:DC%2BB2cXitFCktbzF [DOI: https://dx.doi.org/10.1038/s42003-024-06918-8] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/39358426][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11446919]

35. Liao, H et al. Human NOP2/NSUN1 regulates ribosome biogenesis through non-catalytic complex formation with box C/D SnoRNPs. Nucleic Acids Res.; 2022; 50, pp. 10695-10716.1:CAS:528:DC%2BB3sXmsF2gt78%3D [DOI: https://dx.doi.org/10.1093/nar/gkac817] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36161484][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9561284]