Enhancement of Intracellular Cholesterol Efflux in Chondrocytes Leading to Alleviation of Osteoarthritis Progression

Abstract

Objective: Osteoarthritis (OA) is the most common degenerative disease worldwide, with no practical means of prevention and limited treatment options. Recently, our group unveiled a novel mechanism contributing to OA pathogenesis in association with abnormal cholesterol metabolism in chondrocytes. In this study, we aimed to establish a clinical link between lipid profiles and OA in humans, assess the effectiveness of cholesterol-lowering drugs in suppressing OA development in mice, and uncover the cholesterol-lowering mechanisms that effectively impede OA progression.

Methods: Five clinically approved cholesterol-lowering drugs (fenofibrate, atorvastatin, ezetimibe, niacin, and lomitapide) were injected into the knee joints or administered with diet to mice with OA who underwent destabilization of the medial meniscus induction and were fed a 2% high-cholesterol diet. Gene expression linked to cholesterol metabolism was determined using microarray analysis. Furthermore, the in vivo functions of these genes were explored through intra-articular injection of either its inhibitor or adenovirus.

Results: Logistic regression analysis confirmed a close relationship between the diagnostic criteria of hyperlipidemia based on serum lipid levels and OA incidence. Among the cholesterol-lowering drugs examined, fenofibrate exerted the most significant protective effect against cartilage destruction, which was attributed to elevated levels of high-density lipoprotein cholesterol that are crucial for cholesterol efflux. Notably, cholesterol efflux was suppressed during OA progression via down-regulation of apolipoprotein A1-binding protein (AIBP) expression. Overexpression of AIBP effectively inhibits OA progression.

Conclusion: Our results suggest that restoration of cholesterol homeostasis to a normal state through administration of fenofibrate or AIBP overexpression, both of which induce cholesterol efflux, offers an effective therapeutic option for patients with OA.

Figure 1

Hypercholesterolemia is closely associated with OA pathogenesis, and regulation of cholesterol transport modulates OA progression. (A) Serum TC, HDL, LDL, TC to HDL ratio, and LDL to HDL ratio of human male and female patients with OA diagnosed with KL grades 0 and 3 to 4, respectively (n = 8 per group). *P < 0.05, **P < 0.01. (B) Representative safranin O staining images of knee joint (scale, 100 μm), (C) serum cholesterol profiling, and (D) scoring of OARSI grade, SBP thickness, and osteophyte maturity in male mice fed an RD or HCD with or without BI0115 intra‐articular injection and subjected to DMM surgery (RD, HCD n = 26, HCD+BI0115 n = 18, total 70 mice). Circles represent individual samples; lines with whiskers show the mean ± SEM. *P < 0.05, ***P < 0.005. Significant differences were detected by (A and C) two‐tailed t‐test for serum cholesterol profiling and (D) SBP thickness between sham and DMM and one‐way analysis of variance with Bonferroni post hoc test for SBP thickness among RD, HCD, and HCD+BI0115 or Mann–Whitney U test for OARSI, osteophyte, and synovitis score between sham and DMM and Kruskal–Wallis test for OARSI, osteophyte, and synovitis score among RD, HCD, and HCD+BI0115. DMM, destabilization of the medial meniscus; F, female; HCD, high‐cholesterol diet; HCD+BI0115, HCD with BI0115; HDL, high‐density lipoprotein; KL, Kellgren–Lawrence; LDL, low‐density lipoprotein; M, male; ns, not significant; OA, osteoarthritis; OARSI, Osteoarthritis Research Society International; RD, regular diet; SBP, subchondral bone plate; TC, total cholesterol.

Figure 2

Dietary administration of Chol‐lowering drugs regulates serum Chol levels and fenofibrate inhibits cartilage destruction in male mice with OA who were fed an HCD. (A) Timeline of animal experiment with OA for Chol‐lowering drug administration. (B) Serum Chol profiling (n ≥ 13 per group). *P < 0.05, **P < 0.01, ***P < 0.005. (C) Representative safranin O staining images of knee joint (scale, 100 μm) and scoring of OARSI grade, SBP thickness, and osteophyte maturity in male mice fed an HCD with Chol‐lowering drugs and subjected to DMM surgery (HCD n = 29, HCD with drug administration n = 20, total 129 mice). Circles represent individual samples; lines with whiskers show the mean ± SEM. *P < 0.05, ***P < 0.005. Significant differences were detected by (D) two‐tailed t‐test for SBP thickness between sham and DMM groups and (B) one‐way ANOVA with Bonferroni or Dunnett T3 post hoc test for serum Chol profiling and (D) SBP thickness between groups with HCD and HCD with drug administration or Mann–Whitney U test for OARSI grade and osteophyte maturity between sham and DMM group and Kruskal–Wallis test for OARSI grade and osteophyte maturity between groups HCD and HCD with drug administration. Chol, cholesterol; DMM, destabilization of the medial meniscus; HCD, high‐Chol diet; HCD+Atorvastatin, HCD with atorvastatin; HCD+Ezetimibe, HCD with ezetimibe; HCD+Fenofibrate, HCD with fenofibrate; HCD+Lomitapide, HCD with lomitapide; HCD+Niacin, HCD with niacin; HDL, high‐density lipoprotein; LDL, low‐density lipoprotein; ns, not significant; OA, osteoarthritis; OARSI, Osteoarthritis Research Society International; SBP, subchondral bone plate; TC, total Chol; w w^‐1, weight in weight.

Figure 3

Dietary administration of cholesterol‐lowering drugs regulates serum cholesterol levels, and fenofibrate and ezetimibe inhibit cartilage destruction in female mice with OA fed an HCD. (A) Serum cholesterol profiling (n = 10 per group). ***P < 0.005. (B) Representative safranin O staining images of the knee joint (scale, 100 μm) and scoring of OA parameters including OARSI grade, SBP thickness, and osteophyte maturity in female mice fed an HCD with cholesterol‐lowering drugs and subjected to DMM surgery (n = 10 mice per group, total 60 mice). Circles represent individual samples; lines with whiskers show the mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.005. Significant differences were detected by (C) two‐tailed t‐test for SBP thickness between sham and DMM groups and (A) one‐way ANOVA with Dunnett T3 post hoc test for serum cholesterol profiling and (C) SBP thickness between groups with HCD and HCD with drug administration or Mann–Whitney U test for OARSI grade and osteophyte maturity between groups with sham and DMM and Kruskal–Wallis test for OARSI grade and osteophyte maturity between groups with HCD and HCD with drug administration. DMM, destabilization of the medial meniscus; HCD, high‐cholesterol diet; HCD+Atorvastatin, HCD with atorvastatin; HCD+Ezetimibe, HCD with ezetimibe; HCD+Fenofibrate, HCD with fenofibrate; HCD+Lomitapide, HCD with lomitapide; HCD+Niacin, HCD with niacin; HDL, high‐density lipoprotein; LDL, low‐density lipoprotein; ns, not significantly different; OA, osteoarthritis; OARSI, Osteoarthritis Research Society International; SBP, subchondral bone plate; TC, total cholesterol.

Figure 4

Chol efflux is suppressed in osteoarthritis chondrocytes via HDL oxidization and expression of efflux mediators is reduced. (A) Representative immunofluorescence image of DIO‐oxLDL uptake (n = 4) and (B) heatmap of microarray showing genes of Chol metabolic process (GO 0008203) and HDL particle (GO 0034364) (n = 3) in chondrocytes treated with IL‐1β (2.5 ng/mL) or infected with adenovirus overexpressing Epas1 (Ad‐E) and Zip8 (Ad‐Z). (C) Oxidized HDL levels in culture media incubated with chondrocytes treated with IL‐1β (5 ng/mL) or TNFα (100 ng/mL) in the presence of HDL (10 μg/mL; n = 4). *P < 0.05. (D) mRNA levels of Saa1 and indicated Chol efflux mediators in chondrocytes treated with IL‐1β or TNFα (n = 6). *P < 0.05, **P < 0.01, ***P < 0.005. (E) Cellular Chol levels in Chol‐preloaded chondrocytes (800 μM) treated with IL‐1β (2.5 ng/mL) or TNFα (50 ng/mL) in the presence of HDL (50 μg/mL; n = 3). Circles represent individual samples; lines with whiskers show mean ± SEM. *P < 0.05. Significant differences were detected by (C) two‐tailed t‐test for oxidized HDL enzyme‐linked immunosorbent assay and (E) Chol assay or (D) one‐way analysis of variance with Dunnett T3 post hoc test for mRNA levels. Ad‐C, control adenovirus; Chol, cholesterol; DIO‐oxLDL, 3,3'‐dioctadecyloxacarbocyanine‐lebelled oxidized LDL; GO, Gene Ontology; HDL, high‐density lipoprotein; IL, interleukin; mRNA, messenger RNA; ns, not significant; TNF, tumor necrosis factor.

Figure 5

AIBP is down‐regulated in OA cartilage, and its overexpression promotes cholesterol efflux from OA chondrocytes and inhibits disease progression. (A) Representative AIBP immunostaining images of cartilage from mice with DMM surgery. Scale, 100 μm (n = 4). (B) Cholesterol efflux from chondrocytes infected with a multiplicity of infection of 400 of Ad‐Aibp (n = 5). ***P < 0.005. (C) mRNA levels of indicated genes (n ≥ 4) in chondrocytes infected with Ad‐Aibp in the presence of IL‐1β (2.5 ng/mL). *P < 0.05, **P < 0.01, ***P < 0.005. (D and F) Representative safranin O staining images of knee joint (scale, 100 μm) and (E and G) OARSI grade and SBP thickness in male mice fed a (D and E) regular diet (n = 14 mice per group, total 28 mice) or (F and G) HCD (n = 17 mice per group, total 34 mice) with Ad‐Aibp intra‐articular injection and subjected to DMM surgery. Circles represent individual samples; lines with whiskers show mean ± SEM. *P < 0.05, ***P < 0.005. Significant differences were detected by two‐tailed t‐test for (B) cholesterol efflux and (E and G) SBP thickness or (C) one‐way analysis of variance with Dunnett T3 post hoc test for mRNA levels or (E and G) Mann–Whitney U test for OARSI grade. Ad‐Aibp, adenovirus overexpressing Aibp; Ad‐C, control adenovirus; AIBP, apolipoprotein A1–binding protein; DMM, destabilization of the medial meniscus; HCD, high‐cholesterol diet; HCD+Ad‐Aibp, HCD with Ad‐Aibp; HCD+Ad‐C, HCD with Ad‐C; IL, interleukin; mRNA, messenger RNA; ns, not significant; OA, osteoarthritis; OARSI, Osteoarthritis Research Society International; SBP, subchondral bone plate.