Targeted apoptosis of macrophages and osteoclasts in arthritic joints is effective against advanced inflammatory arthritis

Rheumatoid arthritis (RA) is a chronic autoimmune disease characterized by synovial inflammation, cartilage damage, and bone erosion, which can progress to joint deformity and disability. Synovial macrophages and osteoclasts (OCs) are key mediators of inflammation and bone resorption in RA, respectively, and both express high levels of αvβ3 integrin. In RA, macrophages and OCs exhibit reduced apoptotic rates, contributing to persistent inflammation and joint destruction. Existing therapies (glucocorticoids, cytokine-neutralizing biologics, and JAK inhibitors) have limitations including systemic side effects, infection risk, incomplete response, and limited durability, particularly in advanced disease. The authors hypothesized that selectively inducing apoptosis in both inflammatory macrophages and OCs within arthritic joints would alleviate synovial inflammation and reverse bone erosion in advanced RA. They propose MMP9-responsive, RGD-functionalized PLGA nanoparticles to deliver the pro-apoptotic agent celastrol (CEL) specifically to these cell types in the inflamed joint microenvironment.

The study builds on evidence that synovial macrophage accumulation is an early hallmark of RA and that OCs localize to erosion sites and can also amplify inflammation via cytokine production. αvβ3 integrin is implicated in activated macrophage-driven inflammation and OC-dependent bone resorption. Apoptosis tightly regulates macrophage and OC lifespans; however, in RA their apoptosis is reduced. Standard treatments—glucocorticoids and cytokine-targeting biologics—reduce inflammation but have significant side effects (osteoporosis, infection risk) and limited efficacy in some patients. JAK inhibitors may improve bone outcomes but also carry infection risks. Celastrol is a cytotoxic triterpene with pro-apoptotic and anti-inflammatory properties, previously targeted successfully to mesangial cells to treat glomerulonephritis, improving efficacy and reducing toxicity. RGD peptides target αvβ3 integrins, but integrins are also expressed in normal tissues and neovascular endothelium, raising off-target concerns. MMP9 is upregulated in arthritic joints and can be exploited to cleave protective PEG coatings, enabling site-specific unmasking of RGD ligands for targeted uptake.

• Nanoparticle design: Developed MMP9-cleavable, PEG- and RGD-modified PLGA nanoparticles (PRNPs). RGD-PEG2000-PLGA20000 synthesized via maleimide-thiol coupling (Mal-PEG-PLGA + Cys-RGD). CEL-loaded PLGA NPs (CEL-NPs) and CEL-loaded RGD-PLGA NPs (CEL-RNPs) prepared by emulsion/solvent evaporation. PRNPs formed by conjugating mPEG2000–MMP9-cleavable peptide (GPLGLAGQC) to CEL-RNPs in aqueous conditions.
• Characterization: DLS for size and zeta potential (CEL-NPs ~155.7 nm; CEL-RNPs ~154.1 nm; CEL-PRNPs ~162.2 nm; mildly negative zeta); TEM for morphology (spherical, uniform); high encapsulation efficiency (~90%); serum stability in 10% FBS at 37 °C over 24 h; in vitro release in PBS at 37 °C.
• Cell models: Murine bone marrow macrophages (BMMs) differentiated to OCs with M-CSF + RANKL; LPS-activated macrophages from BMMs; human OCs derived from RA patient PBMCs; human synovial macrophages isolated by MACS from RA synovial tissue; HUVECs ± TNF activation to model angiogenic endothelium.
• Cellular uptake: Coumarin-6 (C6)-loaded NPs, RNPs, PRNPs ± recombinant MMP9; confocal microscopy and flow cytometry in OCs, LPS-activated macrophages, human OCs, human synovial macrophages, and HUVECs.
• Apoptosis assays: Annexin V-FITC/PI flow cytometry after 24 h exposure to CEL-NPs, CEL-RNPs, CEL-PRNPs (100 ng/mL CEL) ± MMP9; JC-1 staining for mitochondrial membrane potential in macrophages.
• Animal model: Adjuvant-induced arthritis (AIA) in male Wistar rats; early stage at day 4, advanced at day 17 post-induction. Also unilateral inflamed joint model via local adjuvant injection.
• Biodistribution: DiD-labeled NPs/RNPs/PRNPs administered IV; in vivo imaging at 2–48 h; ex vivo organ and joint imaging; assessment of selective accumulation in inflamed vs normal joints.
• Joint cell targeting in vivo: Immunofluorescence of joint sections for DiD colocalization with CD68 (macrophages) and CD51 (OCs).
• In vivo efficacy study: AIA rats with advanced arthritis treated IV every 3 days with 1 mg/kg CEL equivalents (CEL solution, CEL-NPs, CEL-RNPs, CEL-PRNPs); anti-TNF (1.5 mg/kg, IP, every 2 days) as comparator; serial measurements of ankle diameter and paw thickness.
• Histology and biomarkers: H&E, safranin O, toluidine blue for cartilage matrix (GAGs); TRAP staining for OCs; immunohistochemistry for CD68, MMP9, RANKL, OPG, OCN, ALP, TNF, IL-1β; ELISAs for serum TNF, IL-1β, RANKL, OPG, TRAP.
• Bone microarchitecture: Ex vivo micro-CT of ankle joints; quantitative analysis of BMD, BS/BV, trabecular number (Tb.N), trabecular thickness (Tb.Th), trabecular separation (Tb.Sp).
• Safety: Histopathology and TUNEL in major organs; serum enzymes; comparison of off-target apoptosis/toxicity across formulations.
• Statistics: Mean ± SD; Student’s two-sided t test for two-group; two-way ANOVA for multiple comparisons; p < 0.05 significant.

• Targeted uptake: RGD-modified NPs (RNPs) showed selective uptake by OCs and LPS-activated macrophages; PRNPs exhibited low uptake without MMP9 but, upon MMP9 exposure, showed high uptake via RGD–αvβ3 interactions. Similar dual targeting observed in human RA-derived OCs and synovial macrophages. PRNPs reduced off-target uptake in activated HUVECs due to PEG shielding until MMP9 cleavage.
• Induction of apoptosis: CEL-RNPs and CEL-PRNPs (with MMP9) significantly increased apoptosis in OCs and activated macrophages versus CEL-NPs or PRNPs without MMP9; JC-1 assays indicated mitochondrial depolarization consistent with apoptosis. Human RA-derived cells mirrored these findings.
• Biodistribution: In AIA rats, PRNPs accumulated preferentially in inflamed joints with extended retention, while reducing distribution to lung, liver, and spleen compared to RNPs, indicating improved selectivity and circulation via cleavable PEG.
• In vivo joint cell targeting: DiD-PRNPs colocalized with CD68+ macrophages and CD51+ OCs in synovium more than NPs or RNPs, supporting in situ PEG cleavage and RGD-mediated targeting.
• Cellular depletion and apoptosis in joints: CEL-PRNPs induced the highest TUNEL+ apoptosis in inflamed joints, predominantly in macrophages and OCs, and markedly reduced TRAP+ OCs and CD68+ macrophages; serum TRAP levels normalized.
• Restored bone remodeling balance: CEL-PRNPs lowered RANKL expression and RANKL/OPG ratio in joints and blood toward normal; increased ALP and OCN-positive osteoblasts, indicating enhanced bone formation.
• Anti-inflammatory effects: CEL-PRNPs yielded the lowest serum and joint levels of TNF and IL-1β among treatments, reflecting effective suppression of inflammatory macrophages.
• Therapeutic outcomes: In advanced arthritis, CEL-PRNPs outperformed CEL solution, anti-TNF, CEL-NPs, and CEL-RNPs in reducing ankle diameter and paw thickness, alleviating synovial hyperplasia, and restoring cartilage GAGs.
• Bone repair: Micro-CT showed CEL-PRNPs reversed bone erosion, increasing BMD and improving trabecular parameters (higher Tb.N and Tb.Th, lower Tb.Sp), with bone surfaces approaching normal.
• Safety: CEL-PRNPs minimized off-target organ accumulation and apoptosis relative to free CEL and RNPs, indicating improved safety profile with negligible side effects at tested doses.

The study demonstrates that insufficient apoptosis of synovial macrophages and OCs contributes to persistent inflammation and bone erosion in advanced RA. By engineering MMP9-cleavable PEG-coated, RGD-functionalized PLGA nanoparticles, the authors achieved selective delivery of celastrol to αvβ3-expressing macrophages and OCs specifically within the inflamed joint microenvironment. This strategy enhances cellular uptake and apoptosis of these pathogenic cells while limiting off-target effects in normal tissues due to PEG shielding outside MMP9-rich inflamed sites. Compared to standard anti-TNF therapy, which is less effective in advanced disease and provides limited bone protection, CEL-PRNPs offer a cell-directed cytotoxic approach that broadly reduces inflammatory mediators and depletes OCs, thereby restoring the RANKL/OPG balance and promoting bone repair. The findings support targeted pro-apoptotic nanotherapy as a means to induce inflammatory remission, reverse bone erosion, and potentially mitigate secondary osteoporosis in advanced RA. The MMP9-responsive design also addresses limitations of RGD-only targeting that can result in non-specific distribution to integrin-expressing normal tissues and neovasculature.

The authors developed MMP9-responsive, RGD-modified PLGA nanoparticles to deliver celastrol selectively to inflammatory macrophages and osteoclasts in arthritic joints. This dual-targeting approach effectively induced apoptosis in both cell types, reduced pro-inflammatory cytokines, restored bone remodeling balance (lower RANKL/OPG), repaired cartilage and bone erosion, and improved clinical signs in a rat model of advanced RA, with minimal off-target toxicity. The work introduces a promising therapeutic strategy for advanced inflammatory arthritis that simultaneously addresses synovial inflammation and bone destruction. Future research should assess long-term safety, pharmacokinetics, and efficacy in additional preclinical models, including αvβ3 integrin and MMP9 knockout mice, and explore translational potential toward clinical trials.

• Off-target integrin expression poses theoretical risk; although mitigated by MMP9-cleavable PEG, comprehensive long-term toxicity and biodistribution studies are needed.
• Mechanistic validation in genetically modified models (αvβ3 integrin knockout, MMP9−/−) was not performed and is proposed for future work.
• Human data are limited to ex vivo/primary cell assays from a small number of late-stage RA patients; in vivo human efficacy and safety remain to be established.
• The study focuses on a single dosing level and schedule; dose-ranging, chronic administration, and combination therapy evaluations are warranted.