Celastrol suppresses neovascularization in rat aortic vascular endothelial cells stimulated by inflammatory tenocytes via modulating the NLRP3 pathway

Rotator cuff injury is prevalent and burdensome, with high re-tear rates after surgery. Tendon-bone healing involves stem cells, growth factors, immune-inflammatory responses, and vascular regeneration. Inflammation is closely linked to tendinopathy, with elevated mediators (e.g., PGE2, IL-1, IL-6) and activation of signaling pathways (IFN, NF-κB, STAT-6). Excessive angiogenesis during repair can exacerbate scarring and impair functional recovery. VEGFA is a central angiogenic factor and is upregulated by inflammatory cytokines (e.g., TNF-α, IL-6), linking inflammation to angiogenesis. Celastrol, a bioactive component from Tripterygium wilfordii, has immunosuppressive and anti-inflammatory properties across diseases, but its role in rotator cuff injury and inflammation-induced angiogenesis is unclear. This study investigates whether celastrol can suppress LPS-induced inflammatory activation in tenocytes, reduce pro-angiogenic signaling (VEGFA), inhibit endothelial tube formation, and improve tendon-to-bone healing in a rat rotator cuff tear (RCT) model, potentially via modulation of the NLRP3 pathway.

The paper contextualizes angiogenesis as a double-edged process in tendon healing: while necessary, excessive neovascularization worsens scarring and mechanical outcomes. VEGFA is highlighted as a key mediator, influenced by inflammatory cytokines and hypoxic factors (e.g., HIF-1α). Prior studies report dynamic VEGF expression during tendon healing, with early peaks after acute injury and context-dependent roles where limiting angiogenesis can improve later healing stages. The NLRP3 inflammasome is implicated in tendon pathology, fibrosis, and regulation of pro-inflammatory cytokines such as IL-1β. Celastrol has documented anti-inflammatory and disease-modifying effects across rheumatoid arthritis, osteoarthritis, autoimmune disease, and cancer, acting through multiple pathways (e.g., NF-κB, TLR2/NF-κB, inhibition of NLRP3 activation). These data underpin the hypothesis that targeting NLRP3-driven inflammation could modulate VEGFA-mediated angiogenesis in tendon healing, and that celastrol may serve as a therapeutic.

In vitro: Primary rat tenocytes were isolated from 5–6-week-old female Sprague-Dawley rats (~100 g). Tendons were minced, explant-adhered, cultured in DMEM + 10% FBS + 1% antibiotic-antimycotic, then passaged 3–4 times. Tenocytes were seeded at 1×10^5/mL in six-well plates, cultured 1 day, and treated with LPS at 0.5 or 1 μg/mL for 6 or 12 h. Conditioned media were collected. RAOECs (rat aortic vascular endothelial cells; YS2272C) were used for tube formation assays on Matrigel-coated 96-well plates; 10,000 cells/well were incubated for 6 h in LPS-tenocyte conditioned media or control media; images were acquired (Zeiss Axio Observer) and quantified with AngioTool. VEGFA secretion in tenocyte supernatants was quantified by ELISA (Beyotime PV960). Celastrol was administered to LPS-treated tenocytes to assess effects on inflammation and VEGFA secretion. Molecular assays: RNA was extracted (TRIzol), cDNA synthesized (TransGen kits), and RT-qPCR performed (TransStart Top Green qPCR SuperMix) using specified primers (IL-1β, TNF-α, Nlrp3, Vegfa, Gapdh). Protein analysis involved RIPA lysis, SDS-PAGE, PVDF transfer, and immunoblotting for NLRP3 (Abcam ab263899), IL-1β (Abcam EPR21086), with GAPDH as loading control; detection by ECL. Immunocytochemistry: tenocytes on poly-L-lysine coverslips were fixed, stained for NLRP3 with AlexaFluor 488 secondary, counterstained with DAPI, and imaged by confocal microscopy (Zeiss LSM 880). In vivo: Rat RCT model: a 1.5 cm anterolateral shoulder incision exposed the supraspinatus; half the tendon was detached from the greater tuberosity and repaired with 4-0 sutures. Groups: Sham, RCT, and RCT + Celastrol (1 mg/kg intra-articular weekly). At 4 or 8 weeks, biomechanical testing (MTS 858) measured ultimate load to failure and stiffness (preload up to 5 N; extension rate 14 mm/s). Tendon tissues were analyzed by RT-qPCR for Vegfa and Nlrp3 mRNA. Statistics: Experiments repeated ≥3 times; comparisons by Student’s t-test (two groups) or one-way ANOVA (≥3 groups); significance at P<0.05. In vivo sample size: three rats per group at each time point. Ethical approval: First Affiliated Hospital, Jinan University (202108246-25).

- LPS induced inflammatory activation in tenocytes: time- and dose-dependent increases in mRNA of NLRP3, TNF-α, IL-1β, and VEGFA at 0.5–1 μg/mL for 6–12 h; NLRP3 and IL-1β proteins also upregulated, with strongest response at 0.5 μg/mL for 12 h (selected for subsequent experiments). - LPS elevated secreted VEGFA in tenocyte conditioned media (ELISA; significant increase vs control), and this conditioned media significantly enhanced RAOEC tube formation (increased total tube length) (P<0.0001 in representative analyses). - Celastrol suppressed LPS-induced inflammation in tenocytes: significant reductions in NLRP3 and IL-1β mRNA and protein (immunoblot), and decreased NLRP3 immunostaining intensity (immunocytochemistry) compared with LPS alone (P<0.05 to P<0.001 depending on marker). - Celastrol reduced VEGFA secretion from LPS-induced tenocytes (ELISA; P<0.001) and inhibited RAOEC angiogenesis driven by inflammatory tenocyte conditioned media (significant reduction in total tube length; P<0.0001). - In RCT rats, celastrol improved biomechanics at 4 and 8 weeks: ultimate load to failure and stiffness were significantly higher in RCT + celastrol vs RCT controls (P<0.01 to P<0.0001), reversing injury-induced impairments. - Tendon tissues from RCT rats showed increased Vegfa and Nlrp3 mRNA, both of which were significantly reduced by celastrol treatment at 4 and 8 weeks (P<0.05 to P<0.0001). Overall, celastrol mitigates inflammation and pro-angiogenic signaling in vitro and improves structural-functional healing in vivo, consistent with modulation of the NLRP3–VEGFA axis.

The study demonstrates that inflammatory activation of tenocytes (via LPS) elevates NLRP3/IL-1β and VEGFA expression and secretion, producing conditioned media that drives endothelial tube formation, implicating tenocyte-derived factors in neovascularization during tendon repair. Celastrol counteracts these effects by downregulating NLRP3 and IL-1β, reducing VEGFA secretion, and consequently inhibiting RAOEC angiogenesis. In vivo, celastrol enhances tendon-to-bone healing after RCT, restoring biomechanical strength and stiffness while lowering Vegfa and Nlrp3 expression in tendon, supporting the mechanistic link between NLRP3-driven inflammation, aberrant angiogenesis, and compromised healing. The findings align with literature indicating that excessive angiogenesis can worsen scarring and mechanical outcomes, and that modulating angiogenic pressure improves longer-term repair. By targeting the NLRP3 pathway, celastrol provides a dual anti-inflammatory and anti-angiogenic approach that may optimize the vascular milieu for tendon-bone integration and functional recovery.

Celastrol suppresses inflammation-induced neovascularization by modulating the NLRP3 pathway in tenocytes, reducing VEGFA secretion and endothelial tube formation in vitro, and improves biomechanical healing in a rat RCT model while lowering Vegfa and Nlrp3 expression in vivo. These results identify celastrol as a potential anti-angiogenic adjunct to promote tendon-to-bone healing. Future work should elucidate broader signaling networks affected by celastrol (e.g., via transcriptomics/proteomics), profile a wider range of inflammatory mediators and secretome components, and develop delivery strategies to overcome celastrol’s solubility and bioavailability limitations while ensuring safety for longer-term use.

- Mechanistic depth is limited: downstream pathways beyond NLRP3/IL-1β (e.g., collagen production, ECM organization, cytoskeletal dynamics) were not interrogated; omics approaches (RNA-seq/proteomics) are needed. - Inflammatory profiling was incomplete: not all pro- and anti-inflammatory mediators (e.g., IL-2, IL-6, IL-10) or secreted factors beyond VEGFA were assessed. - Pharmacological constraints of celastrol: poor water solubility, low bioavailability, narrow therapeutic window, potential adverse and immunosuppressive effects; long-term safety and optimized delivery require further evaluation. - Sample sizes were modest (n=3 per group per timepoint in vivo), which may limit generalizability and statistical power for some endpoints.