Efficacy of pH-Responsive Surface Functionalized Titanium Screws in Treating Implant-associated S. aureus Osteomyelitis with Biofilms Formation

Abstract

Implant-associated Staphylococcus aureus (S. aureus) osteomyelitis (IASO) leads to high orthopedic implant failure rates due to the formation of Staphylococcal abscess community within the bone marrow and bacterial colonization in the osteocyte lacuno-canalicular network (OLCN). To address this, antimicrobial peptides (HHC36)-loaded titania nanotubes (NTs) are developed on titanium screws (Ti-NTs-P-A), which integrate pH-responsive polymethacrylic acid to control HHC36 release for eradicating bacteria in IASO. Colony-forming unit assay confirmed that Ti-NTs-P-A screws maintained sustainable antibacterial effectiveness, killing over 65% of S. aureus even after multiple bacterial solution replacements. Notably, Ti-NTs-P-A screws exhibit significant pH-responsive HHC36 release behavior and bactericidal activity, consistent with the phenotype of peptides-killed bacteria from scanning electron microscopy. Transcriptome sequencing results reveal that Ti-NTs-P-A screws interfered with ribosome formation and disrupted the arginine biosynthesis, which is crucial for bacterial survival in acidic environments. In the non-infected implant model, the bone-implant contact ratio of the Ti-NTs-P-A screw is 2.3 times that of the clinically used titanium screw. In an IASO model, Ti-NTs-P-A screws effectively eradicated bacteria within the OLCN, achieving an 80% infection control rate and desirable osteointegration. Collectively, Ti-NTs-P-A screws with pH-responsive antibacterial properties exhibit great potential for eradicating bacteria and achieving osseointegration in IASO.

Keywords: antimicrobial peptides; biofilm; implant‐associated S. aureus osteomyelitis; osteointegration; pH‐responsive.

Scheme 1

Design of the titanium screws modified with pH‐responsive TiO₂ nanotubes for on‐demand release of antimicrobial peptides and enhanced implant osseointegration in implant‐associated S. aureus osteomyelitis.

Figure 1

The preparation and pH‐activated AMPs (HHC36) release of the functionalized titanium screws. A) Schematic illustrating the preparation process of functionalized titanium screws, which were modified with PMAA‐gated and AMPs‐loading TiO₂ nanotubes. B) SEM images and C) pore size distribution of the TiO₂ nanotubes on the Ti‐NTs screws. D) The SEM image of a cross‐sectional view of TiO₂ nanotubes with a compact oxide layer. E) Curves of acoustic signal and frictional force during the scratch test. F) High‐resolution XPS N 1s and G) Ti 2p spectra of the indicated screws exhibited the immobilization of dopamine and PMAA molecules. H) AMPs loading content of the Ti‐NTs‐A and Ti‐NTs‐P‐A screws (n = 3). I) HHC36 released from the Ti‐NTs‐A and Ti‐NTs‐P‐A screws (n = 3), where the pH of the PBS solution was adjusted from 7.4 to 5.0 for the 96–144 h period. Specifically, the average HHC36 release rates 48 h before and after pH change (from 7.4 to 5.0) were denoted as V_a1 and V_a2 in the Ti‐NTs‐A group, and as V_b1 and V_b2 in the Ti‐NTs‐P‐A group, respectively. J) Illustration of the mechanism of pH‐activated AMPs release of the Ti‐NTs‐P‐A screw. At pH 7.4, PMAA molecules swell, limiting AMPs release by obstructing the nanotubes. When the pH value drops to 5.0, PMAA molecules collapse to open the nanotubes, rapidly releasing AMPs. Data are shown as the mean ± SD, n.s. represents no significance. AMPs, antimicrobial peptides; PMAA, polymethacrylic acid; NHS, N‐hydroxysuccinimide; EDC, 1‐ethyl‐3‐(3‐dimethylaminopropyl)‐carbodiimide hydrochloride.

Figure 2

In vitro and ex vivo evidence of pH‐activated antibacterial properties of the functionalized titanium screws. A) The antibacterial activity of the indicated screws against S. aureus cultures for 1h, and for four cycles (n = 3). B) The antibacterial activity of the soaking solution of the indicated screws at pH 5.0 and pH 7.4 against S. aureus, was assessed by CFU (n = 3). C,D) Representative 3D reconstructed fluorescence images of live/dead stained S. aureus after incubation with the indicated screws for 4h (Green, live, and dead bacteria; Red, dead bacteria). E) The friction stabilities of functionalized screws were tested by screwing them into the bone, unscrewing them, and then examining their morphology and function. The friction due to implantation did not affect the physical properties of Ti‐NTs‐P‐A, as shown by representative SEM images in F) (top: screw body; bottom: screw tip). G) The HHC36 release from the unscrewed Ti‐NTs‐A and Ti‐NTs‐P‐A screws (n = 3), where the pH of the PBS solution was adjusted from 7.4 to 5.0 for the 96–144 h period. Specifically, the average HHC36 release rates 48 h before and after the pH change (from 7.4 to 5.0) were denoted as V_a3 and V_a4 in the Ti‐NTs‐A group and denoted as V_b3 and V_b4 in the Ti‐NTs‐P‐A group, respectively. H) The antibacterial activity of the soaking solution of the unscrewed screws at pH 5.0 and pH 7.4 against S. aureus (n = 3). The data are shown as the means ± SDs, ^*** p < 0.001, in comparison with the Ti‐NTs group, ^### p < 0.001 in comparison with the Ti‐NTs‐A group.

Figure 3

Bacterial membrane disruption caused by the functionalized screws. SEM images of S. aureus after cultured with the functionalized screws at A) pH 7.4 or B) pH 5.0 for 4 h. Red arrows indicate normal‐growing S. aureus with round and smooth appearance. Yellow arrows indicate the dead bacteria with deformed or ruptured membranes.

Figure 4

Transcriptome analysis of the S. aureus after cultured with or without the Ti‐NTs‐P‐A screw for 3 h. A) Schematic diagram of the RNA sequence analysis of the Ti‐NT‐P‐A and control groups. B) Heatmap representing the top 100 genes significantly differentially expressed between the Ti‐NT‐P‐A group and the control group. C) Volcano plot of differentially expressed genes. The green dots and red dots indicate upregulated and downregulated differentially expressed genes, respectively. The grey dots represent nonsignificantly expressed genes. D) GO annotations and E) KEGG enrichment analyses of differentially expressed genes (p value < 0.05). Heatmap of the differentially expressed genes related to F) ribosome (top 10 DEGs) and G) arginine biosynthesis.

Figure 5

In vitro and in vivo biocompatibility and osseointegration evaluation of the functionalized titanium screws. A,B) Representative western blot images of the expression of osteogenesis‐related proteins and C,D) corresponding quantitative analyses (n = 3). E) Immunofluorescence images of RUNX2 and COL1 staining of MEFs in different groups. F) Schematic diagram of the osseointegration and biocompatibility test of functionalized screws in an uninfected bone defect model. G) The implantation process of functionalized screws in rabbit tibial plateau. H) The micro‐CT 3D reconstructed images and I) bone tissue volume/total tissue volume (BV/TV) analysis from micro‐CT results of the newly formed bone surrounding the screws after 56 days of implantation (n = 3). J) The methylene blue‐basic magenta staining and corresponding K) bone‐implant contact (BIC) ratio analysis of the non‐decalcified bone sections for osseointegration assay (n = 3). Green arrows indicate gaps between the screw and bone tissue. Blue arrows indicate bone‐implant contact areas. In I,J), The data are shown as the means ± SDs, ^* p < 0.05, ^** p < 0.01, ^*** p < 0.001.

Figure 6

Radiographic and bacteriological assay of the therapeutic effect of the functionalized titanium screws in implant‐associated osteomyelitis. A) Schematic illustration of a rabbit 1‐stage revision model for implant‐associated S. aureus osteomyelitis. A Ti screw was contaminated in S. aureus suspension overnight, air‐dried for 30 min, and implanted in a rabbit's left tibia. 7 days after the implantation, the screw was removed followed by debridement and irrigation, and a sterile screw with or without functionalized surface was implanted for another 28 days. B) Representative images of harvested rabbit tibia 28 days post‐revision. White arrows indicated the purulent mass. C) The in vivo antibacterial activity of the indicated screws via CFU assay (n = 5). D) The reconstructed micro‐CT images of rabbit tibia 28 days post‐revision. Blue arrows indicate the infectious bone destruction of subchondral bone. Yellow arrows indicated the infectious bone destruction and collapse of the growth plate of the tibia. E) The micro‐CT 3D reconstructed images and corresponding F) BV/TV analysis of the newly formed bone surrounding the screws (n = 3). The data are shown as the means ± SDs, ^* p < 0.05, ^** p < 0.01, ^*** p < 0.001.

Figure 7

Histology illustrates the therapeutic effect of the functionalized titanium screws in implant‐associated osteomyelitis. A) Representative images of H&E staining of decalcified rabbit tibiae. Black arrows indicate inflammatory cells. B) Representative images of Gram staining of decalcified rabbit tibiae. Red arrows indicated Gram‐positive bacteria. C) Representative images of methylene blue‐basic magenta staining of undecalcified bone sections. Yellow arrows indicate the bone‐implant contact area. D) The BIC ratios of indicated screws after 28 days of implantation (n = 3). E) Histologic outcomes of indicated screws after 28 days of implantation. F) Diagram of the proposed mechanism by which functionalized titanium screws (Ti‐NTs‐P‐A) exhibit pH‐responsive antibacterial properties for bacterial eradication and osteointegration in implant‐associated osteomyelitis. The data are shown as the means ± SDs, ^* p < 0.05, ^** p < 0.01.