Human Infrapatellar Fat Pad Mesenchymal Stem Cell–derived Extracellular Vesicles Purified by Anion Exchange Chromatography Suppress Osteoarthritis Progression in a Mouse Model

Background Extracellular vesicles derived from mesenchymal stem cells (MSCs) show great promise in treating osteoarthritis (OA). However, studies from the perspective of clinical feasibility that consider an accessible cell source and a scalable preparation method for MSC-extracellular vesicles are lacking. Questions/purposes (1) Does an infrapatellar fat pad obtained from patients undergoing TKA provide a suitable source to provide MSC-extracellular vesicles purified by anion exchange chromatography? Using an in vivo mouse model for OA in the knee, (2) how does injection of the infrapatellar fat pad–derived MSC-extracellular vesicles alter gait, cartilage structure and composition, protein expression (Type II collagen, MMP13, and ADAMTS5), subchondral bone remodeling and osteophytes, and synovial inflammation? Methods The infrapatellar fat pad was collected from three patients (all female; 62, 74, 77 years) during TKA for infrapatellar fat pad–derived MSC culturing. Patients with infection, rheumatic arthritis, and age > 80 years were excluded. MSC-extracellular vesicles were purified by anion exchange chromatography. For the animal study, we used 30 male C57BL/6 mice aged 10 weeks, divided into six groups. MSC-extracellular vesicles were injected weekly into the joint of an OA mouse model during ACL transection (ACLT). To answer our first research question, we characterized MSCs based on their proliferative potential, differentiation capacity, and surface antigen expression, and we characterized MSC-extracellular vesicles by size, morphology, protein marker expression, and miRNA profile. To answer our second research question, we evaluated the effects of MSC-extracellular vesicles in the OA mouse model with quantitative gait analysis (mean pressure, footprint area, stride length, and propulsion time), histology (Osteoarthritis Research Society International Score based on histologic analysis [0 = normal to 24 = very severe degeneration]), immunohistochemistry staining of joint sections (protein expression of Type II collagen, MMP13, and ADAMTS5), and micro-CT of subchondral bone (BV/TV and Tb.Pf) and osteophyte formation. We also examined the mechanism of action of MSC-extracellular vesicles by immunofluorescent staining of the synovium membrane (number of M1 and M2 macrophage cells) and by analyzing their influence on the expression of inflammatory factors (relative mRNA level and protein expression of IL-1β, IL-6, and TNF-α) in lipopolysaccharide-induced macrophages. Results Infrapatellar fat pads obtained from patients undergoing TKA provide a suitable cell source for producing MSC-extracellular vesicles, and anion exchange chromatography is applicable for isolating MSC-extracellular vesicles. Cultured MSCs were spindle-shaped, proliferative at Passage 4 (doubling time of 42.75 ± 1.35 hours), had trilineage differentiation capacity, positively expressed stem cell surface markers (CD44, CD73, CD90, and CD105), and negatively expressed hematopoietic markers (CD34 and CD45). MSC-extracellular vesicles purified by anion exchange chromatography had diameters between 30 and 200 nm and a typical cup shape, positively expressed exosomal marker proteins (CD63, CD81, CD9, Alix, and TSG101), and carried plentiful miRNA. Compared with the ACLT group, the ACLT + extracellular vesicle group showed alleviation of pain 8 weeks after the injection, indicated by increased area (0.67 ± 0.15 cm 2 versus 0.20 ± 0.03 cm 2 , -0.05 [95% confidence interval -0.09 to -0.01]; p = 0.01) and stride length (5.08 ± 0.53 cm versus 6.20 ± 0.33 cm, -1.12 [95% CI -1.86 to -0.37]; p = 0.005) and decreased propulsion time (0.22 ± 0.06 s versus 0.11 ± 0.04 s, 0.11 [95% CI 0.03 to 0.19]; p = 0.007) in the affected hindlimb. Compared with the ACLT group, the ACLT + extracellular vesicles group had lower Osteoarthritis Research Society International scores after 4 weeks (8.80 ± 2.28 versus 4.80 ± 2.28, 4.00 [95% CI 0.68 to 7.32]; p = 0.02) and 8 weeks (16.00 ± 3.16 versus 9.60 ± 2.51, 6.40 [95% CI 2.14 to 10.66]; p = 0.005). In the ACLT + extracellular vesicles group, there was more-severe OA at 8 weeks than at 4 weeks (9.60 ± 2.51 versus 4.80 ± 2.28, 4.80 [95% CI 0.82 to 8.78]; p = 0.02), indicating MSC-extracellular vesicles could only delay but not fully suppress OA progression. Compared with the ACLT group, the injection of MSC-extracellular vesicles increased Type II collagen expression, decreased MMP13 expression, and decreased ADAMTS5 expression at 4 and 8 weeks. Compared with the ACLT group, MSC-extracellular vesicle injection alleviated osteophyte formation at 8 weeks and inhibited bone loss at 4 weeks. MSC-extracellular vesicle injection suppressed inflammation; the ACLT + extracellular vesicles group had fewer M1 type macrophages than the ACLT group. Compared with lipopolysaccharide-treated cells, MSC-extracellular vesicles reduced mRNA expression and inhibited IL-1β, IL-6, and TNF-α in cells. Conclusion Using an OA mouse model, we found that infrapatellar fat pad–derived MSC-extracellular vesicles could delay OA progression via alleviating pain and suppressing cartilage degeneration, osteophyte formation, and synovial inflammation. The autologous origin of extracellular vesicles and scalable purification method make our strategy potentially viable for clinical translation. Clinical Relevance Infrapatellar fat pad–derived MSC-extracellular vesicles isolated by anion exchange chromatography can suppress OA progression in a mouse model. Further studies with large-animal models, larger animal groups, and subsequent clinical trials are necessary to confirm the feasibility of this technique for clinical OA treatment.