Neurovascularization inhibiting dual responsive hydrogel for alleviating the progression of osteoarthritis

Abstract

Treating osteoarthritis (OA) associated pain is a challenge with the potential to significantly improve patients lives. Here, we report on a hydrogel for extracellular RNA scavenging and releasing bevacizumab to block neurovascularization at the osteochondral interface, thereby mitigating OA pain and disease progression. The hydrogel is formed by cross-linking aldehyde-phenylboronic acid-modified sodium alginate/polyethyleneimine-grafted protocatechuic acid (OSAP/PPCA) and bevacizumab sustained-release nanoparticles (BGN@Be), termed OSPPB. The dynamic Schiff base bonds and boronic ester bonds allow for injectability, self-healing, and pH/reactive oxygen species dual responsiveness. The OSPPB hydrogel can significantly inhibit angiogenesis and neurogenesis in vitro. In an in vivo OA model, intraarticular injection of OSPPB accelerates the healing process of condyles and alleviates chronic pain by inhibiting neurovascularization at the osteochondral interface. The injectable hydrogel represents a promising technique to treat OA and OA associated pain.

Fig. 1. The injectable, self-healing, pH/ROS dual-responsive, polycationic hydrogel reversed TMJOA and alleviated chronic pain by inhibiting neurovascularization at osteochondral interface.

Be bevacizumab, VEGF vascular endothelial growth factor, exRNA extracellular ribonucleic acid, VEGFR VEGF receptor. Created in BioRender.

Fig. 2. Preparation and characterization of the OSPPB hydrogel.

a The facile fabrication route of the OSPPB hydrogel. b The gelation, injectability, and self-healing property of the OSPPB hydrogel. c ATR-FTIR of PPCA, OSAP, OSPP, and OSPPB, showing the Schiff base bonds and boronic ester bonds leading to gelation. d SEM of OSPP and OSPPB, showing the uniformly distributed BGN@Be on the surface of OSPPB hydrogel. The red circle represents the BGN. Scale bars = 200 μm (left) and 5 μm (right). e SEM-EDS of OSPPB hydrogel (areas with BGN distribution), showing the Ca, P and Si elements on OSPPB hydrogel. f Frequency-sweep, Time-sweep sequence, and g Shear-thinning test of the OSPP and OSPPB hydrogels. G’ means storage modulus and G” means loss modulus.

Fig. 3. pH/ROS dual responsiveness, degradation, antioxidation property and positive electrical property of the OSPPB hydrogel.

a–d Images, frequency-sweep, and SEM (Scale bars = 200 μm) of the OSPPB hydrogel after incubation with PBS (pH = 5) and/or H₂O₂ (n = 6, F(3, 20) = 140.7, p < 0.0001). e Degradation of OSPP and OSPPB hydrogels in PBS. f Characteristics of the S concentration changes over 14 days. g Photographs of DPPH after culture with OSPPB for 30 min, with the corresponding statistical results (n = 6, F(4, 25) = 567.7, p < 0.0001). h Zeta potential of OSPP and OSPPB hydrogels. i–k Characteristics of the binding test, showing the binding between OSPPB and RNA (n = 6, F(2, 15) = 930.3, p < 0.0001). Created in BioRender. Scale bar = 50 μm. l Molecular dynamic modulation and its quantification for RNA (light blue), VEGF (red) and PEI (Polyethyleneimine, purple). Data are shown as the means and standard deviations. One-way ANOVA is followed by the Tukey-Kramer method for post hoc multiple comparisons.

Fig. 4. The OSPPB hydrogel inhibits angiogenesis and neurogenesis.

a, b Representative microscopy images of EPCs and TG cells showing cell migration, tube formation, and morphology after 24 h of treatment. Scale bars = 100 μm. c, d Quantification of migrated cells and the migration distance of EPCs after treatment. n = 6, F(6, 35) = 45.9, p < 0.0001 in (c). n = 6, F(6, 35) = 89.54, p < 0.0001 in (d). e, f Quantification of the interactions and dendritic length of TG cells after treatment. n = 6, F(6, 35) = 39.66, p < 0.0001 in (e). n = 6, F(6, 35) = 85.86, p < 0.0001 in (f). Data are shown as the means and standard deviations. One-way ANOVA is followed by the Tukey-Kramer method for post hoc multiple comparisons.

Fig. 5. The OSPPB hydrogel alleviates TMJOA in mice.

a The facile fabrication route depicting the design and the results of the in vivo experiment. Reproduced under the terms of a Creative Commons Attribution license (CC-BY-4.0). b Illustrative macroscopic views above the condyles at 3 weeks after intra-articular injection, depicting comparative morphology in the control (CON) and experimental (unilateral anterior crossbite, UAC) groups. c Representative images of H&E staining, SF staining, sliver staining, SEM (arrows indicate micro-fractures), and micro-CT. d–k OARSI score and semi-statistical analysis of the images in (c). n = 6, H(3) = 16.12, p = 0.0011 in (d). n = 6, F(3, 20) = 28.63, p < 0.0001 in (e). n = 6, F(3, 20) = 43.48, p < 0.0001 in (f). n = 6, H(3) = 14.56, p = 0.0022 in (g). n = 6, H(3) = 14.92, p = 0.0019 in (h). n = 6, F(3, 20) = 27.57, p < 0.0001 in (i). n = 6, F(3, 20) = 16.42, p < 0.0001 in (j). n = 6, F(3, 20) = 38.22, p < 0.0001 in (k). l qRT-PCR analysis of the gene expression of neurovascular factors (Vegf, Pdgfb, Ngf, Ntn1, Ntn3, Ntn4) in the condyles. Scale bars = 200 μm (b), 100 μm (c1–3), 10 μm (c4), and 500 μm (c5). Data are shown as the means and standard deviations. Kruskal–Wallis (KW) test is followed by Dunn’s test for post hoc multiple comparisons in (d, g, h). One-way ANOVA is followed by the Tukey-Kramer method for post hoc multiple comparisons in (e, f, i–l).

Fig. 6. The OSPPB hydrogel ameliorates changes at the osteochondral interface caused by TMJOA.

a Mechanical characterization of the TMJ osteochondral interface by AFM at 3 weeks after different treatments in the CON and UAC groups. b SEM-EDS micrographs of osteochondral interface. Scale bars = 20 μm.

Fig. 7. Celecoxib and OSPPB hydrogel exhibit remarkable pain-alleviating functionality.

a Representative tracks of mice at 3 weeks in the CON and UAC groups, as documented in the EPM test, with the corresponding statistical results (c–e). One-way ANOVA is followed by the Tukey-Kramer method for post hoc multiple comparisons in (c, d). n = 6, F(3, 20) = 16.83, p < 0.0001 in (c). n = 6, F(3, 20) = 3.696, p = 0.0289 in (d). n = 6, H(3) = 11.6, p = 0.0089 in (e). b Representative tracks of mice in the OFT, with the corresponding statistical results (f–h). One-way ANOVA accompanied by the Tukey-Kramer method is used for post hoc multiple comparisons. n = 6, F(3, 20) = 19.4, p < 0.0001 in (f). n = 6, F(3, 20) = 31.85, p < 0.0001 in (g). n = 6, F(3, 20) = 13.68, p < 0.0001 in (h). i Results of the von Frey test revealing the pain relief after different treatments (n = 6, H(3) = 10.21, p = 0.0168). Reproduced under the terms of a Creative Commons Attribution license (CC-BY-4.0). j EEG spectrograms of the stimulus before and after small brush stimulation. Created in BioRender. Data are shown as the means and standard deviations. One-way ANOVA is followed by the Tukey-Kramer method for post hoc multiple comparisons in (c, d, f–h). Kruskal–Wallis (KW) test is followed by Dunn’s test for post hoc multiple comparisons in (e, i).

Fig. 8. The OSPPB hydrogel effectively attenuates neurovascularization at the osteochondral interface.

a Representative images of the distribution of exRNA at the osteochondral interface of condyles in the CON and UAC groups at 3 weeks after different treatments. b Representative images of the trigeminal ganglion anterograde tracing and immunofluorescence staining of nerves (PGP9.5 and CGRP), vessels (CD31), and VEGF. c Representative images of the distribution of pain-related factors (COX2, DCC, and SP). d The semi-statistical analysis of (a). e–h The semi-statistical analysis of (b). i–k The semi-statistical analysis of (c). n = 6, F(3, 20) = 213.6, p < 0.0001 in (d). n = 6, F(3, 20) = 66.07, p < 0.0001 in (e). n = 6, F(3, 20) = 146.4, p < 0.0001 in (f). n = 6, F(3, 20) = 96.29, p < 0.0001 in (g). n = 6, F(3, 20) = 57.17, p < 0.0001 in (h). n = 6, F(3, 20) = 130.5, p < 0.0001 in (i). n = 6, F(3, 20) = 159.9, p < 0.0001 in (j). n = 6, F(3, 20) = 143.9, p < 0.0001 in (k). Scale bars = 50 μm (left in a), 10 μm (right in a), 30 μm (b, c). Data are shown as the means and standard deviations. One-way ANOVA is followed by the Tukey-Kramer method for post hoc multiple comparisons.