Nanoscale ZnO doping in prosthetic polymers mitigate wear particle-induced inflammation and osteolysis through inhibiting macrophage secretory autophagy

Abstract

Wear particles produced by joint replacements induce inflammatory responses that lead to periprosthetic osteolysis and aseptic loosening. However, the precise mechanisms driving wear particle-induced osteolysis are not fully understood. Recent evidence suggests that autophagy, a cellular degradation process, plays a significant role in this pathology. This study aimed to clarify the role of autophagy in mediating inflammation and osteolysis triggered by wear particles and to evaluate the therapeutic potential of zinc oxide nanoparticles (ZnO NPs). We incorporated ZnO into the prosthetic material itself, ensuring that the wear particles inherently carried ZnO, providing a targeted and sustained intervention. Our findings reveal that polymer wear particles induce excessive autophagic activity, which is closely associated with increased inflammation and osteolysis. We identified secretory autophagy as a key mechanism for IL-1β secretion, exacerbating osteolysis. Both in vitro and in vivo experiments demonstrated that ZnO-doped particles significantly inhibit autophagic overactivation, thereby reducing inflammation and osteolysis. In summary, this study establishes secretory autophagy as a critical mechanism in wear particle-induced osteolysis and highlights the potential of ZnO-doped prosthetic polymers for targeted, sustained mitigation of periprosthetic osteolysis.

Keywords: Aseptic loosening; Macrophages; Osteoclastogenesis; Periprosthetic osteolysis; Secretory autophagy; Wear particle-induced inflammation.

Graphical abstract

Scheme 1

Graphical abstract of the current study. (A)Macrophages at the implant site and those migrating from distant locations phagocytose wear particles (such as PEEK and PE), leading to the secretion of inflammatory cytokines such as IL-1β. These cytokines then promote osteoclast activation and bone resorption, contributing to periprosthetic osteolysis. (B) In macrophages, wear particles cause lysosomal damage, leading to the activation of caspase, which converts pro-IL-1β into mature IL-1β (mIL-1β). This mIL-1β is then secreted extracellularly through secretory autophagy, mediated by TRIM16 and Sec22b. ZnO-containing wear particles inhibit this secretory autophagy pathway, thereby reducing the secretion of IL-1β.

Fig. 1

Fabrication and characterization of the wear debris. (A) Schematic illustration of materials preparation. The average tensile strength (B) for the PEEK, PEEK-ZnO, PE, and PE-ZnO groups was 79.20 MPa, 77.37 MPa, 35.63 MPa, and 32.11 MPa, respectively, while the average elastic modulus (C) was 3.901 GPa, 4.127 GPa, 1.642 GPa, and 1.868 GPa, respectively. (D) The stress-strain curves of different groups. (E) The average weight content distribution of the Zinc element in the PEEK-ZnO and PE-ZnO groups was 11.87 % and 8.705 %, respectively. Each bar represents the average of three independent samples. (F) SEM images of the fabricated materials. (G) SEM images of the wear particles. (H) EDS mapping and spectrum of different elements in different wear particles. FT-IR (I) and XRD (J) patterns of the wear particles. Broad XPS spectrum of the fully scanned region (K) and the (L) Zn 2p region of the wear particles.

Fig. 2

Proteomic profiling of synovial fluid in Labrador following THA using PEEK-HXLPE prostheses. (A) The implanted PEEK-HXLPE prostheses post-operation immediately. (B–C) The joint capsule and synovial fluid 18 months after the operation. (D) SEM images of the PEEK and PE surfaces before and after the implantation. (E) Microarray heat map visualizing the fold change in expression of proteins. (F) Volcano plots show the expression levels of proteins in different groups. (G) KEGG enrichment analysis result. (H) GO enrichment analysis results classified into biological process (BP), cell composition (CC), and molecular function (MF). (I) Results of protein-protein interaction (PPI) network.

Fig. 3

Assessment of the wear particle-induced inflammation and osteoclastogenesis. (A,B) CCK-8 assay of the RAW264.7 cells co-cultured with wear particles (0, 12.5, 25, 50, 100, 200 and 400 μg/mL) for 12 h and 24 h (n = 3). (C) Quantification of IL-1β protein levels secreted into the cell culture supernatant, as measured by ELISA. All experiments were performed with n = 3 independent biological replicates. (D) Schematic illustration of the double-chamber co-culturing system. (E, F) TRAP and F-actin staining after 7 days of co-culturing (scale bar: 100 μm). (G) Bone slice resorption after 10 days of co-culturing (scale bar: 100 μm). (H) The number of TRAP⁺ osteoclasts. (I,J) Quantitative analysis of the F-actin area and nuclei numbers of the osteoclasts. (K) Quantitative analysis of the bone resorption area in different groups. (L) Western blot analysis of osteoclastogenesis-related proteins, including NFATc1, CTSK, and TRAP, in BMDMs from each group. (M − O) RT-PCR results of osteoclastogenesis-related genes, including NFATc1, CTSK, and TRAP, in BMDMs from each group. Data are presented as mean ± SD (n = 3). (* indicates P < 0.05, ** indicates P < 0.01, *** indicates P < 0.001, and **** indicates P < 0.0001).

Fig. 4

Autophagy is triggered and mediates the secretion of IL-1β upon stimulation of wear particles. (A, B) WB and immunofluorescence results of autophagy-related proteins in RAW264.7 cells challenged with different wear particles (scale bar: 20 μm). (C) Quantification of LC3-II level in WB. (D) Quantification of LC3 immunofluorescence intensity. (E) Concentrations of IL-1β and pro-IL-1β in culture supernatants of RAW264.7 cells stimulated with wear particles for 12 h in the absence or presence of 3-methyladenine (3-MA) (5 mM). (F) Immunofluorescence staining on RAW264.7 cells and 3D reconstruction for IL-1β (red), LC3 (green), and DAPI (blue) (scale bar: 5 μm). Yellow spots indicate co-localization. (G) 3D co-localization analysis was conducted using the IMARIS coloc plugin. (H) Co-localization analysis of red channel (IL-1β) and green channel (LC3) conducted by FIJI software. (I) Colocalization line tracing analysis from images in (F). Gray arrows indicate the region of red-green overlap. (J) Pearson's colocalization coefficient for IL-1β and LC3. Sample size: n = 3 per group. All experiments were performed with n = 3 independent biological replicates. (ns indicates P > 0.05, *** indicates P < 0.001, and **** indicates P < 0.0001). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 5

ZnO doping inhibits the autophagy-mediated IL-1β secretion. (A) Immunofluorescence staining on RAW264.7 cells for IL-1β (red), LC3 (green), and DAPI (blue) after treated with different wear particles (scale bar: 5 μm). (B, C) Quantification of IL-1β and LC3 immunofluorescence intensity. (D) Mander's co-localization coefficient of LC3 and IL-1β. (E) Concentrations of IL-1β in culture supernatants of RAW264.7 cells stimulated with different wear particles. Sample size: n = 3 per group. All experiments were performed with n = 3 independent biological replicates. (** indicates P < 0.01, *** indicates P < 0.001, and **** indicates P < 0.0001). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 6

Sec22b and TRIM16 are involved in autophagy-mediated IL-1β secretion. (A) Immunofluorescence staining on RAW264.7 cells and 3D reconstruction for TRIM16 (red), LC3 (green), and DAPI (blue) (scale bar: 5 μm). Yellow spots indicate co-localization. (B) 3D co-localization analysis conducted by IMARIS coloc plugin. (C) Co-localization analysis of red channel (TRIM16) and green channel (LC3) conducted by FIJI software. (D) Colocalization line tracing analysis from images in (A). Gray arrows indicate the region of red-green overlap. (E) Pearson's colocalization coefficient for TRIM16 and LC3. (F) Immunofluorescence staining and 3D reconstruction for Sec22b (purple), LC3 (green), and DAPI (blue) (scale bar: 5 μm). White spots indicate co-localization. (G) 3D co-localization analysis conducted by IMARIS coloc plugin. (H) Co-localization analysis of purple channel (TRIM16) and green channel (LC3) conducted by FIJI software. (I) Colocalization line tracing analysis from images in (F). Gray arrows indicate the region of purple-green overlap. (J) Pearson's colocalization coefficient for Sec22b and LC3. Sample size: n = 3 per group. All experiments were performed with n = 3 independent biological replicates. (*** indicates P < 0.001). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 7

Autophagic flux and inflammation assessment in vivo. (A) Schematic illustration of intra-articular injection of prosthetic wear particles. (B) WB result of autophagy flux in normal synovial cells treated with Torin (50 nM) and Torin (50 nM) + CQ (50 μM) for 4 h. (C) Immunofluorescence staining on normal synovial cells for LC3 (green), ATG16L (cyan), FIP200 (blue), and WIPI2 (red) (scale bar: 20 μm). (D) WB results of autophagy-related proteins in the periprosthetic tissue lysates. (E) IHC staining results for LC3 of the periprosthetic tissue section (scale bar: 25 μm). (F, G) Quantification of LC3-II and p62 level in WB result (D). (H) IHC quantification for LC3. (I, J) Quantification of fluorescence intensity of LC3 and ATG16L. (K, L) Immunofluorescence staining on the isolated synovial cells for LC3 and ATG16L after treatment of Torin (50 nM) and Torin (50 nM) + CQ (50 μM) for 4 h (scale bar: 50 μm). (M, N) IHC staining and quantification results for IL-1β of the periprosthetic tissue section (scale bar: 25 μm). Sample size: n = 3 per group. All experiments were performed with n = 3 independent biological replicates. (* indicates P < 0.05, ** indicates P < 0.01, *** indicates P < 0.001, and **** indicates P < 0.0001). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 8

Protective effects of ZnO doping on wear particle-induced mice calvaria osteolysis. (A) Schematic illustration of wear particle-induced mice calvaria osteolysis model. (B–E) Micro-CT analysis and relevant parameters for wear particles-induced osteolysis in mice calvaria (scale bar: 100 μm). Analysis was performed with n = 3 independent experiments. (F) H&E staining in different groups (scale bar: 100 μm). (G) TRAP staining in different groups (scale bar: 100 μm; arrows represent TRAP-positive cells). (H) The number of TRAP-positive cells in different groups. (I–J) Osteolysis score of different groups. (* indicates P < 0.05, ** indicates P < 0.01, *** indicates P < 0.001, and **** indicates P < 0.0001).