Wogonin pretreatment of infrapatellar fat pad mesenchymal stem cell-derived exosomes advances articular cartilage repair in osteoarthritis

Abstract

Aims

Osteoarthritis (OA) is a chronic joint disorder characterized by progressive cartilage degeneration, inflammation, and subchondral bone remodelling. Herein, the role of exosomes (Exos) extracted from wogonin-pretreated infrapatellar fat pad mesenchymal stem cells (MSCs^IPFP) was explored, and their ability to promote cartilage defect repair in OA was clarified.

Methods

In this study, the therapeutic effects of wogonin on MSCs^IPFP-derived Exos and OA chondrocytes were investigated in vitro, and a mouse OA model was studied in vivo. Human-derived chondrocytes and MSCs^IPFP were isolated and cultured. These cells were characterized through morphological observation, toluidine blue staining, immunofluorescence staining, detection of stem cell surface markers, and induction of directed differentiation. DiI dye was used to label and trace MSCs^IPFP-derived Exo (MSCs^IPFP-Exo). Chondrocyte inflammation and the mouse OA model were induced using interleukin (IL)-1β and destabilization of the medial meniscus (DMM) surgery. To evaluate chondrocyte proliferation and apoptosis, cell counting kit (CCK)-8 assay and flow cytometry were conducted. Articular cartilage destruction in mice was assessed using haematoxylin and eosin (H&E) staining, Safranin O/Fast Green staining, and the Osteoarthritis Research Society International (OARSI) score. Additionally, immunohistochemical staining and/or western blot were performed to examine the expression of Sox9, aggrecan, type II collagen (collagen II), a disintegrin and metalloproteinase with thrombospondin motifs 5 (ADAMTS5), and matrix metalloproteinase 13 (MMP13).

Results

The isolated chondrocytes could uptake MSC^IPFP-Exo. Wogonin-MSC^IPFP-Exo enhanced chondrocyte proliferation and suppressed apoptosis; Wogonin-MSC^IPFP-Exo significantly alleviated cartilage tissue damage in OA mice compared to untreated controls and MSC^IPFP-Exo-treated OA mice in in vivo experiments. Mechanistically, wogonin-MSC^IPFP-Exo upregulated Sox9, aggrecan, and collagen II protein levels, while downregulating ADAMTS5 and MMP13 protein levels compared to untreated controls and MSC^IPFP-Exo-treated OA mice.

Conclusion

Wogonin pretreatment significantly enhances the ability of MSC^IPFP-Exo to promote cartilage defect repair, and it is expected to be a promising agent for the clinical treatment of OA. Further preclinical and clinical studies are necessary to validate the safety, efficacy, and long-term outcomes of this therapeutic approach before its translation into clinical practice for the treatment of OA.

Cite this article: Bone Joint Res 2025;14(12):1064–1079.

Article focus

• The therapeutic effect of exosomes (Exos) extracted from wogonin-pretreated infrapatellar fat pad mesenchymal stem cells (MSCs^IPFP-Exo) on osteoarthritis (OA) mouse model.

Key messages

• Wogonin-MSC^IPFP-Exo enhanced chondrocyte proliferation and suppressed apoptosis.

• Wogonin-MSC^IPFP-Exo significantly alleviated cartilage tissue damage in OA mice.

• Wogonin significantly enhanced Exo biogenesis in MSC^IPFP and affected Exo-carried microRNA levels.

Strengths and limitations

• Wogonin pretreatment significantly enhances the ability of MSC^IPFP-Exo to promote cartilage defect repair, and it is expected to be a promising agent for the clinical treatment of OA.

• In vivo evaluation of safety and duration of drug action of wogonin-MSC^IPFP-Exos is lacking.

Introduction

Osteoarthritis (OA) is a degenerative joint disease caused by a variety of factors, with pathological features that include degenerative destruction of articular cartilage, synovitis, sclerosis or cystic degeneration of the subchondral bone, hyperosteogeny of the joint margins, and osteophyte formation.¹ OA onset is linked to many factors including age, trauma, inflammation, obesity, and strain.² With the population getting older and the proportion of obese people increasing, OA incidence is increasing, and the incidence rate of the age group over 65 years is as high as 80%, and OA has become the fourth most disabling disease in the world.³ Early OA can be treated conservatively with oral anti-inflammatory and analgesic drugs or intra-articular injection of sodium vitrate and glucocorticoids to relieve symptoms. However, patients with advanced OA suffer from severe pain and joint deformity, with massive destruction of articular cartilage. Symptoms can no longer be controlled by oral medication, and most of them need to undergo joint arthroplasties, which have a series of drawbacks and shortcomings including high costs, limited durability of prostheses, and the risk of complications like infections and implant loosening.⁴ There are no effective drugs and means to curb the progression and exacerbation of the disease and eradicate the suffering of patients.

The core problem of OA is articular cartilage damage and degeneration. Articular cartilage mainly consists of hyaline cartilage, which is difficult for the body to regenerate and repair effectively due to the lack of blood vessels and nerves. The stem cell-based regenerative medicine has attracted much attention owing to the development of tissue engineering. Exosomes (Exos) are considered to be one of the crucial secretory products of mesenchymal stem cells (MSCs) and are key messengers for intercellular communication.⁵ Compared with MSCs, Exo is a non-cellular component that enriches the active components of the source cells, has less immunogenicity, and also overcomes the tumorigenic potential of MSCs after transplantation, which is promising for clinical application. Among them, infrapatellar fat pad-derived mesenchymal stem cells (MSCs^IPFP) exhibit superior chondrogenic potential, higher proliferation rates, and stronger immunomodulatory properties compared to bone marrow-derived MSCs.^6,7 These advantages make IPFP an ideal source for joint-related therapies, including OA treatment. For example, the part of the infrapatellar fat pad that needs to be removed in knee arthroplasty has less damage to the donor, which makes it an ideal donor choice.⁸

Wogonin (5,7-dihydroxy-8-methoxyflavone) is one of the most active flavonoids found in various plants, such as the roots of Scutellaria baicalensis, which has long been used in traditional Chinese medicine to treat inflammatory diseases⁹ and exhibits significant anti-inflammatory¹⁰ and antioxidant properties.¹¹ Wogonin attenuates articular cartilage damage in arthritic animals via repressing the NF-κB signalling activation.^12,13 Together, these studies illustrate the underlying therapeutic effect of wogonin on OA, and its functions in MSCs-Exo are currently unknown.

In this study, we found that wogonin pretreatment significantly enhanced the production of Exos from MSCs^IPFP. In addition, wogonin pretreatment of MSC^IPFP-Exo could promote chondrocyte proliferation and suppress apoptosis in vitro, and could alleviate cartilage tissue damage in vivo. Mechanistically, wogonin-pretreated MSC^IPFP-Exo upregulated Sox9, type II collagen (collagen II) and proteoglycan (aggrecan) protein levels, and downregulated a disintegrin and metalloproteinase with thrombospondin motifs 5 (ADAMTS5) and matrix metalloproteinase 13 (MMP13) protein levels. These findings provide a new rationale for the wogonin treatment of OA.

Methods

Ethics statement

Ethical approval for human sample collection was obtained from the Institutional Ethics Committee Review Board of the Second Xiangya Hospital, under approval number 2021Yan739. Written informed consent was obtained from all patients prior to sample collection, in accordance with the Declaration of Helsinki¹⁴ and institutional guidelines.

Ethical approval for animal experiments was granted by the Animal Ethics Committee of the Second Xiangya Hospital, under approval number 2022437. All animal procedures were performed in compliance with the ARRIVE guidelines and institutional standards for the care and use of laboratory animals.

Cell isolation, culture, and identification

MSCs^IPFP isolation and culture: The IPFP tissues from trauma patients who underwent amputation surgery in the Second Xiangya Hospital were collected and the adipose tissues were stored in a solution containing phosphate-buffered saline (PBS) (with antibiotics) at 4°C. Tissues were obtained and cut in a sterile environment, supplemented with 0.1% collagenase (Sigma-Aldrich, USA), and digested in a shaker at room temperature (RT) for two hours, and then filtered using 40 μm cell filters. After adding an appropriate amount of MSC growth medium, the tissues were centrifuged (700 g × 10 mins) for the removal of the supernatant. Next, the bottom precipitate was resuspended using MSC growth medium, followed by incubation at 37°C with 5% CO₂, and the non-adherent cells were observed and discarded after 24 hours. The fluid was changed in two to three days.¹⁵ The isolated MSCs^IPFP were identified using flow cytometry (CD29, CD44 and CD90 positive, CD45, CD31 and CD117 negative; all the antibodies were purchased from BioLegend, USA). The adipogenetic, chondrogenic, and osteogenic differentiation abilities of MSCs^IPFP were determined as previously described.^16,17

Human primary chondrocyte isolation and culture: Cartilage tissues were collected from patients undergoing surgery for OA, and samples of articular cartilage from the medial side of patients’ tibial cartilage, including both severely and slightly damaged regions, were cut by removing the attached tissues around the joints in a sterile environment. Subsequently, the cut cartilage slices were rinsed three times with PBS (with antibiotics). After that, the OA cartilage tissues were cut into slices smaller than 1 mm³ and subjected to fiv to six hours of digestion with 0.15% collagenase II (Invitrogen, Thermo Fisher Scientific, USA) at 37°C, during which slices were stirred every 20 minutes for two hours.

Chondrocytes were identified as previously described.¹⁸ After centrifugation, chondrocytes were harvested and cultivated for five to seven days in Dulbecco’s Modified Eagle Medium/Nutrient Mixture F-12 (DMEM-F12) added with 10% fetal bovine serum (FBS) and antibiotics for subsequent experiments. For the identification of chondrocytes, toluidine blue staining, immunofluorescence (IF) staining of collagen type II alpha 1 chain (COL2A1) was evaluated using an Olympus CKX53 fluorescence Microscope (Olympus, Japan) observation. For investigation, OA chondrocytes from passages 2-4 were chosen.

Oil Red O staining

MSCs^IPFP underwent three weeks of adipogenic differentiation by adipogenic differentiation medium (OriCell, China), and was further stained using the Oil Red O Staining Kit (G1262; Solarbio, China) to observe the level of intracellular lipid accumulation. Briefly, after three washings in PBS, cells were fixed with 10% formalin for 20 minutes at RT. Then, after five minutes of washing in 60% isopropanol, cells were subjected to 20 minutes of staining at RT using Oil Red O staining solution. Subsequently, after PBS washing, the excess dye was removed, and the nuclei of the cells were subjected to two minutes of staining with haematoxylin. An inverted microscope was applied to observe and photograph the staining.

Alcian blue staining

MSCs^IPFP after three weeks of induction in a chondrogenic induction medium (OriCell) were stained using an Alcian Blue Staining Kit (G1565; Solarbio). Briefly, differentiated MSCs^IPFP were fixed with formaldehyde for five minutes. Next, cells were rinsed two times in PBS, followed by 50 minutes of staining at RT with drops of Alcian staining solution. Subsequently, cells were rinsed thrice in PBS (1 min each). After PBS washing, the excess dye was removed, and an inverted microscope was applied to observe and photograph the staining.

Alizarin red staining

MSCs^IPFP after three weeks of induction in the osteogenic induction medium (OriCell) were stained using the Osteoblast Calcified Nodule Staining Kit (Alizarin Red S method) (C0148S; Beyotime, China). Briefly, differentiated MSCs^IPFP were subjected to 20 minutes of fixing at RT with 10% formalin. Subsequently, after three PBS washings (5 mins each), cells were stained with an appropriate amount of Alizarin Red S staining solution for 30 minutes at RT. After PBS washing, the excess dye was removed, and an inverted microscope was applied to observe and photograph the staining.¹⁹

Extraction and characterization of Exo from MSC^IPFP

MSCs^IPFP were inoculated in 10 cm dishes and the cell culture media was replaced with Exo-specific serum-free media (41210ES50; Yeasen, China). The cell supernatant of MSC^IPFP was collected upon 24-hour incubation with the addition of wogonin (25 μM; MedChemExpress, Shanghai, China),^20-22 after continuing the culture for two days in Exo-specific serum-free media. The cell suspension was subjected to ten minutes of centrifugation (300 g, 4°C); the rest cells and cell debris were then removed and the supernatant was harvested. Subsequently, the supernatants were transferred to a sterilized centrifuge tube and subjected to ten minutes of centrifugation (2,000 g, 4°C). Then, the supernatants were subjected to 30 minutes of centrifugation (10,000 g, 4°C). The supernatants were subsequently subjected to 90 minutes of ultracentrifugation (100,000 g, 4°C). The supernatant was discarded and the precipitates (isolated Exos) were re-suspended by cooled PBS and subjected to another 90 minutes of ultracentrifugation (100,000 g, 4°C). Finally, precipitates (isolated Exos) were re-suspended by cooled PBS and kept at -80°C for further experiments. Exos (2 μg) on the basis of protein contents determined using Pierce BCA protein assay kit (Thermo Fisher Scientific, USA) were used for incubation with 2 × 10⁵ chondrocytes for 24 hours. MSC^IPFP-Exo were isolated using the same protocol without the wogonin pretreatment.

A ZetaView Nanoparticle Tracking Microscope PMX-120 (Particle Metrix, Germany) was employed to assess MSC^IPFP-Exo size distribution. Transmission electron microscopy (TEM) was used to identify Exo morphology. Exosomal surface marker proteins including CD81 (ab79559, 2 µg/ml), CD9 (ab236630, 1:1,000), CD63 (ab134045, 1:5,000), and Calnexin (ab92573, 1:20,000) (all from Abcam, UK) were analyzed by western blotting.

Nanoparticle tracking analysis

Exo size distribution was determined using the ZetaView Nanoparticle Tracking Microscope PMX-120 (Particle Metrix). The camera was focused and the instrument was calibrated using polystyrene microspheres of 100 nm.

DiI tracing of Exos

Labelling: DiI was supplemented to MSC^IPFP-Exo and Wogonin-MSC^IPFP-Exo at a final concentration of 5 μM, followed by 30 minutes of incubation at RT. The unbound dye from the precipitate was centrifuged (10,000 g, 4°C) for 30 minutes. The precipitate was centrifuged (120,000 g, 4°C) for 70 minutes and then re-suspended in pre-cooled sterile PBS before being kept at low temperature without light exposure.

Tracing: 2 μg/ml DiI-labelled Exos were added into the complete medium and filtered by 0.22 μm filter, followed by 12 hours of co-incubation with chondrocytes. Three PBS washings were carried out repeatedly to remove the unbound free Exos. An Olympus CKX53 fluorescence Microscope was employed to observe DiI-labelled Exo uptake by chondrocytes.

Western blot

Protein samples were obtained by lysing Exos, mouse knee cartilage tissues, and human chondrocytes using radioimmunoprecipitation assay (RIPA) lysate (Beyotime). Protein concentration was detected using the Bicinchoninic Acid (BCA) kit (Beyotime). After the addition and complete mixture of the corresponding protein volume into the sampling buffer (Beyotime), five minutes of heating in a boiling water bath was conducted to denature the protein. The proteins were subjected to electrophoresis (80 V, 30 mins) until the bromophenol blue was migrated into the gel; next, a higher voltage (120 V) was used for one to two hours. The membrane transfer was performed in an ice bath, with a current of 220 mA for 120 minutes. Next, after one to two minutes of washing, membranes were blocked for 60 minutes at RT, and incubated with the primary antibody (glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (5174S, 1:1,000; Cell Signaling Technology (CST), USA), MMP13 (AF5355, 1:1,000; Affinity Bioscience, China), collagen II (ab34712, 1:1,000; Abcam), aggrecan (ab3778, Abcam; ab313636, 1:1,000, Abcam), Sox9 (ab185966, 1:1,000; Abcam), ADAMTS5 (DF13268, 1:500; Affinity Bioscience), β-catenin (AF6266, 1:500; Affinity Bioscience), p-mammalian target of rapamycin (mTOR) (AF3308, 1:500; Affinity Bioscience), and mTOR (AF6308, 1:500; Affinity Bioscience) at 4°C overnight and then with a secondary antibody (horseradish peroxidase-conjugated goat anti-rabbit or anti-mouse immunoglobulin G (IgG), 1: 5,000; Proteintech) for one hour at RT. A chemiluminescent imaging system (Tanon-5200; Tanon, China) was applied to perform detection after a drop of the enhanced chemiluminescence (ECL) substrates (Merck Millipore, USA) to the membrane.

CCK-8 assay

The cell counting kit (CCK)-8 (Beyotime) was employed under the manufacturer’s protocols to evaluate cell viability. Chondrocytes were seeded in a 96-well plate overnight and then incubated with IL-1β and/or Exos. Subsequently, the chondrocytes were continued to be incubated for 0, 24, 48, and 72 hours, then 10 μl of CCK-8 solution was supplemented to each well, followed by incubation for two hours at 37°C. The absorbance was subsequently measured at 450 nm using a microplate reader (Bio-Rad, USA).

Cell apoptosis assay

The Annexin V-FITC/PI apoptosis kit (KeyGen Biotech, China) was employed as per the protocols to evaluate IL-1β- and/or Exo-treated chondrocyte apoptosis, and the assay was carried out immediately after the staining was completed. In short, after being harvested and rinsed in pre-cooled PBS, 1 × 10⁵ to 10⁶ cells were re-suspended in 500 μl of 1× binding buffer. Then, 5 μl of FITC and 5 μl of propidium iodide (PI) dye were added into each tube and mixed well, followed by a 15-minute incubation at RT devoid of light, which was detected as soon as possible after incubation.

IF staining

MSC^IPFP-Exo-, Wogonin-MSC^IPFP-Exo-, and IL-1β-treated chondrocytes were fixed using 4% paraformaldehyde solution for 30 minutes at RT, rested using 0.1% Triton X-100 solution for 20 minutes at RT, and then blocked using 10% goat serum for 30 minutes at RT. Afterward, slides were incubated with collagen II (ab34712, 1:500; Abcam) and MMP13 (ab39012, 1:100; Abcam) (diluted with 10% goat serum; Zhongshan Golden Bridge, China) overnight at 4°C and Alexa Fluor 488-conjugated secondary antibody (diluted with 10% goat serum) for two hours at 37°C. The nuclei were stained for one minute using 4',6-diamidino-2-phenylindole (DAPI), followed by visualization using a fluorescent inverted microscope²³ (green light wavelength of 543 nm and blue light wavelength of 458 nm) at 200×.

Establishment of OA mouse models

Healthy male specific pathogen-free (SPF)-grade eight-week-old C57BL/6J mice (n = 24) were procured from SLAC laboratory animal company (China). All mice were acclimatized and fed for one week before formal experiments. OA was induced in mice using medial meniscus destabilization (DMM) of the right and left knees. After intraperitoneal injection with ketamine and xylazine, mice were anaesthetized, and the skin of the operative area was prepared and disinfected with iodophor. Mice were positioned supine, the joint capsule located on the medial side of the right and left knee patellar tendon was incised, and the medial meniscus-tibial ligament (MTL) was dissected using surgical scissors under the microscope. Subsequently, the joint capsule, the medial vastus muscle, and the connective tissues were sequentially sutured with absorbable sutures, and finally the skin of the operative area was sutured with nonabsorbable sutures, and the knees were sterilized with iodophor. Postoperatively, mice were given ampicillin to prevent infection. For the sham mice, only the knee capsule was opened and the medial MTL did not get dissected.

The experimental mice were randomized into four groups (n = 6 mice): the Sham group, OA group, OA + MSC^IPFP Exos group, and OA + Wogonin-MSC^IPFP-Exos group. After four weeks of DMM surgery, 10 μl of Exo suspension containing 1 × 10¹⁰ Exos particles²⁴ derived from MSC^IPFP or wogonin pretreated MSC^IPFP (Wogonin-MSC^IPFP) was injected into right and left knee joint cavities of the OA + MSC^IPFP Exos group and the OA + Wogonin-MSC^IPFP-Exos group (twice per week for four weeks), respectively, and the sham and OA groups received semiweekly intra-articular injection with an equal amount of PBS into the joint cavity for four weeks. A suspension containing 1 × 10¹⁰ Wogonin-MSC^IPFP-Exos particles was used for each intra-articular injection in the in vivo experiments. Mice were kept in controlled conditions (25°C; 60% relative humidity) under regular light/dark cycles (12 hr light:12 hr dark), during which all mice had free access to a normal chow diet and water. Mice were euthanized at four weeks postoperatively. Both knee joints were collected, with the left knee fixed for histological analysis and the right knee cartilage used for protein expression detection.

Histological scoring systems

The pathological histological changes in cartilage tissues were evaluated using the Mankin and OARSI scoring systems.²⁵ The Mankin score was employed to evaluate the severity of cartilage degeneration, including structural damage, cellularity, and matrix loss.²⁶ The OARSI scoring system quantified cartilage destruction on a scale of 0 to 6, assessing the maximal score. The histological scoring was performed independently by two experienced pathologists (see Acknowledgements) in a blinded manner to ensure objective and reliable evaluation.

H&E staining

After decalcifying, dehydrating, and permeabilizing, the left knee tissues were embedded with paraffin and cut into 4 μm slices. Paraffin slices were deparaffinized using xylene and rehydrated with an increasing concentration of ethanol, followed by staining with a haematoxylin and eosin (H&E) staining kit (Beyotime). Briefly, sections were subjected to ten minutes of immersion in haematoxylin staining solution and five seconds of immersion in hydrochloric acid ethanol for five seconds after tap water washing. After tap water washing again, slices were subjected to ten minutes of immersion in eosin staining (in 70%, 80%, and 90% ethanol for 10 s sequentially), and then ten seconds of immersion in anhydrous ethanol. Lastly, slices were subjected to ten minutes of permeabilization with xylene and mounted by neutral resin, and the microscope was applied to capture the histopathological alterations in the knee joints.

Safranin O/Fast Green staining

Paraffin slices were deparaffinized and dehydrated, followed by staining with the Safranin O/Fast Green Staining Kit (G1053-100ML; Servicebio, China). Briefly, the sections were subjected to five minutes of staining with Fast Green staining solution, and then tap water was used to remove the excess staining solution. Next, slices were subjected to a reaction with 1% hydrochloric acid for 15 seconds and five seconds of staining with Safranin O staining solution, prior to rapid dehydration with anhydrous ethanol, which was repeated four times, each time for three to five seconds. The slices were permeabilized for five minutes with xylene and mounted by neutral resin. Areas of articular cartilage in mice were visualized under an optical microscope (Olympus CKX53 fluorescence Microscope).

Immunohistochemical staining

After being dewaxed and rehydrated, the slices were subjected to antibody incubation following antigen repair, endogenous peroxidase blocking, and non-specific binding site closure, respectively. The primary antibodies were collagen II (ab34712, 1:50; Abcam) and MMP13 (ab39012, 1:200; Abcam). Slices were incubated overnight, and then the Zhongshan Golden Bridge pv-9000 Universal 2-Step Assay Kit and diaminobenzidine (DAB) colour development solution were employed for the subsequent procedures.

Retention ability of exosomes in vivo

MSC^IPFP or MSC^IPFP-derived Exos were incubated with Cy5 NHS ester dye (final concentration 5 μM) for two hours in the dark. Then, the Cy5-labelled Exos were injected into the right knee joint cavities of normal C57BL/6J mice. An in vivo imaging system (IVIS Spectrum; PerkinElmer, USA) was used to observe the in vivo fluorescence signal at day 0 (after injection), day 3, and day 5.

Quantitative real-time polymerase chain reaction

Total RNA from MSC^IPFP or MSC^IPFP-derived Exo was extracted using the TRIzol reagent (Invitrogen) followed by DNase I (Invitrogen) treatment according to the manufacturer’s instructions. Complementary DNA (cDNA) synthesis was performed using oligo (dT) 20 or microRNA (miRNA) reverse transcription primers and Superscript II reverse transcriptase (Invitrogen). Messenger RNA (mRNA) or miRNA expression was determined using SYBR green PCR Master Mix (Qiagen, Germany), with GAPDH or U6 as an endogenous control. Data were processed using the 2^-ΔΔCT method. The primer sequence is listed in Supplementary Table i.

Statistical analysis

All data were presented in the form of mean (SD). Comparisons among groups were assessed using one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test. Comparisons between groups were assessed using independent-samples two-tailed Student’s t-tests. p < 0.05 was considered a statistically significant difference.

Results

Characterization of human primary chondrocytes and MSC^IPFP

Human articular cartilage-derived primary chondrocytes were observed to be long spindle and polygonal in shape under an inverted microscope (Figure 1a). Toluidine blue staining and IF staining were performed to identify cells. Toluidine blue staining was conducted to stain the proteoglycans in the chondrocyte cytoplasm in purple (Figure 1b). Type II collagen is one of the important components of the cartilage matrix in the organism. The primary chondrocytes expressed strong fluorescence in the cytoplasm (Figure 1c), demonstrating that positive fluorescent staining of type II collagen was mainly localized in the cytoplasm. These results indicate that the cells tested are consistent with chondrocyte characteristics.

Fig. 1.

Characterization of human primary chondrocytes and infrapatellar fat pad mesenchymal stem cells (MSCs^IPFP). Human primary chondrocytes were isolated and extracted: a) the microscope (scale bar: 100 μm) was used to observe cell morphology; b) Toluidine blue staining (scale bar: 50 μm) was applied to stain the proteoglycans in the cytoplasm of chondrocytes in purple; c) type II collagen immunofluorescence staining and 4',6-diamidino-2-phenylindole (DAPI) immunofluorescence staining were combined, and chondrocyte nuclei were stained blue by DAPI, whereas type II collagen was stained green (scale bar, 20 μm). Human IPFP-derived cells were isolated and extracted: d) the microscope (scale bar: 100 μm) was applied to observe cell morphology; e) CD29, CD44, CD90, CD31, CD117, and CD45 expression was tested by flow cytometry; f) adipogenic, chondrogenic, and osteogenic differentiation assays (scale bar: 100 μm).

The extracted human MSCs^IPFP were seen to have a typical long shuttle shape and adherent growth (Figure 1d). Flow cytometry results showed that the extracted cell surface markers CD29, CD44, and CD90 were positive (> 70%), while CD31, CD117, and CD45 were negative (< 5%, Figure 1e). Following three weeks of adipogenic differentiation induction, lipid droplet formation was distinctly observed in the cytoplasm using Oil Red O staining. After three weeks of chondrogenic differentiation induction, Alcian blue staining highlighted abundant mucopolysaccharides and cartilage matrix surrounding the cells. For osteogenic differentiation, Alizarin red staining conducted after four weeks revealed scattered calcium nodules and calcified stroma, confirming the successful differentiation into osteoblasts (Figure 1f). It suggests that MSCs^IPFP have multidirectional differentiation potential and could be elicited to differentiate into lipogenic, chondrogenic, and osteogenic cells, respectively.

Identification of MSC^IPFP -Exo

Exos were isolated by collecting cell culture supernatants from untreated MSC^IPFP and wogonin-pretreated MSC^IPFP cultured for 24 hours and ultracentrifuged. The cytotoxicity of wogonin was first examined. Under 0, 12.5, or 25 μM wogonin treatment, cell viability was not affected; however, 50 μM wogonin significantly suppressed cell viability compared with 25 μM wogonin (Figure 2a). Therefore, 25 μM wogonin was used for the following experiments. The number and morphology of Exos were observed using TEM, which showed a disc-shaped double membrane structure, while wogonin treatment significantly increased the number of MSC^IPFP-Exo (Figure 2b). Western blot results suggested that MSC^IPFP-Exo and Wogonin-MSC^IPFP-Exo groups showed positive expression of the Exo-specific surface proteins CD81, CD63, and CD9, and negative expression of Calnexin (Figure 2c). In addition, nanoparticle tracking analysis (NTA) revealed that the particle size of the vesicles ranged from 40 to 200 nm (Figure 2d). Taken together, these results can confirm that the isolated extracellular vesicular material is Exo and that wogonin treatment significantly increases the number of MSC^IPFP-Exo.

Fig. 2.

Identification of infrapatellar fat pad mesenchymal stem cells (MSCs^IPFP)-Exo. a) Cells were treated with 0, 12.5, 25, or 50 μM wogonin and examined for cell viability using cell counting kit (CCK)-8 assay. b) Results of transmission electron microscopy (TEM). c) The expression of exosome-specific surface proteins (CD81, CD63, CD9, and Calnexin) was determined using Western blot. d) Results of nanoparticle tracking analysis (NTA). Exo, exosome.

Wogonin-MSC^IPFP-Exo suppresses chondrocyte apoptosis

Exos were labelled using DiI dye, and then chondrocytes were incubated with DiI-labelled MSC^IPFP-Exo and Wogonin-MSC^IPFP-Exo, respectively, for 12 hours. The fluorescence inverted microscope was applied to observe the chondrocyte changes, and it was found that DiI-labelled Exos had entered into the chondrocytes, which were mainly distributed within the cytoplasm (Figure 3a). The effect of MSC^IPFP-Exo on chondrocyte function was further investigated. CCK-8 results demonstrated that MSC^IPFP-Exo and Wogonin-MSC^IPFP-Exo had no significant impact on normal chondrocyte proliferation (Figure 3b). Cellular inflammation model was induced using IL-1β treatment of chondrocytes to mimic the OA microenvironment in vitro. It was observed that IL-1β treatment considerably suppressed proliferation (Figure 3b) and enhanced apoptosis (Figure 3c) of chondrocytes, whereas MSC^IPFP-Exo and Wogonin-MSC^IPFP-Exo dramatically attenuated the effects of IL-1β upon the cells, with Wogonin-MSC^IPFP-Exo being more effective. In summary, wogonin treatment enhances the promotion of chondrocyte proliferation and inhibition of apoptosis by MSC^IPFP-Exo.

Fig. 3.

Wogonin-MSC^IPFP-Exo suppresses chondrocyte apoptosis. a) Fluorescent photographs of chondrocyte uptake of DiI-labelled Exo (scale bar: 20 μm). The chondrocytes were subjected to 12-hour incubation with DiI-labelled MSC^IPFP-Exo and Wogonin-MSC^IPFP-Exo, respectively. b) Chondrocytes were incubated with MSC^IPFP-Exo or Wogonin-MSC^IPFP-Exo in presence or absence of interleukin (IL)-1β (10 ng/ml) for 24, 48, and 72 hours, and cell counting kit (CCK)-8 assay was carried out to examine changes in chondrocyte viability. c) Flow cytometry was performed to measure chondrocyte apoptosis after 24-hour incubation with Exo or IL-1β (n = 3). Data were analyzed using one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test. **p < 0.01 compared to control; ##p < 0.01 compared to IL-1β group; &&p < 0.01 compared to IL-1β + MSC^IPFP Exo group. Exo, exosome; MSC^IPFP, infrapatellar fat pad mesenchymal stem cells; OD, optical density.

Wogonin-MSC^IPFP-Exo inhibits IL-1β-caused extracellular matrix degradation in chondrocytes

To further clarify the relationship between Wogonin-MSC^IPFP-Exo and OA pathogenesis, we assessed the protein levels of the essential transcription factor Sox9,²⁷ cartilage extracellular matrix (ECM) (collagen II and aggrecan), and catabolic enzymes (MMP13 and ADAMTS5)²⁸ involved in OA pathogenesis. The findings disclosed that IL-1β significantly decreased Sox9, aggrecan, and collagen II levels and increased ADAMTS5 and MMP13 levels within chondrocytes, whereas MSC^IPFP-Exo markedly reversed IL-1β induced ECM degradation, and Wogonin-MSC^IPFP-Exo showed better effect on chondrocytes compared with MSC^IPFP-Exo (Figures 4a to 4e). Together, these results imply that Wogonin-MSC^IPFP-Exo inhibits IL-1β-triggered chondrocyte ECM degradation.

Fig. 4.

Wogonin-MSC^IPFP-Exo suppresses interleukin (IL)-1β-triggered chondrocyte extracellular matrix (ECM) degradation. Chondrocytes were incubated with MSC^IPFP-Exo or Wogonin-MSC^IPFP-Exo in the presence or absence of IL-1β (10 ng/ml) for 24 hours. a) and b) Type II collagen (collagen II) and c) and d) matrix metalloproteinase 13 (MMP13) levels within chondrocytes (scale bar: 20 μm) were assessed using immunoflourescence (IF) staining. e) Sox9, proteoglycan (aggrecan), collagen II, a disintegrin and metalloproteinase with thrombospondin motifs 5 (ADAMTS5), and MMP13 levels within chondrocytes were evaluated using Western blot (n = 3). Data were analyzed using one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test. **p < 0.01 compared to control; #p < 0.05 and ##p < 0.01 compared to IL-1β group; &p < 0.05 and &&p < 0.01 compared to IL-1β + MSC^IPFP Exo group. Exo, exosome; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; MSC^IPFP, infrapatellar fat pad mesenchymal stem cells.

Wogonin-MSC^IPFP-Exos promotes cartilage repair in OA mice

To investigate the effects of Wogonin-MSC^IPFP-Exos upon cartilage in vivo, the in vivo joint retention assay was performed first and confirmed that the Cy5-labelled Exo still existed at Day 5 after intra-articular injection in normal mice (Supplementary Figure a). Then, we established an OA mouse model, injecting MSC^IPFP-Exos and Wogonin-MSC^IPFP-Exos into the joint cavity, and observed the cartilage damage in mice (Figure 5a). After four weeks of treatment, H&E and Safranin O/Fast Green staining showed that the OA mice articular cartilage surface was uneven and accompanied by multiple defects and thinning of the cartilage layer (Figures 5b and 5c). However, after the injection of MSC^IPFP-Exos and Wogonin-MSC^IPFP-Exos, cartilage surface defects were repaired to some extent, the articular surface was smoother, and the cartilage defects were repaired better in Wogonin-MSC^IPFP-Exo treated mice (Figures 5b and 5c). The trends of Mankin and OARSI scores further confirmed the histopathological observations (Figure 5d). It is suggested that Wogonin-MSC^IPFP-Exos accelerate cartilage repair in OA mice.

Fig. 5.

Wogonin-MSC^IPFP-Exos accelerate cartilage repair in osteoarthritis (OA) mice. a) Flowchart of intra-articular injection time. Four weeks after destabilization of the medial meniscus (DMM) surgery, OA mice received intra-articular injection with MSC^IPFP-Exos or Wogonin-MSC^IPFP-Exos twice per week for four weeks. b) Haematoxylin and eosin (H&E) and c) Safranin O/Fast Green (scale bar: 50 μm) were performed to stain the knee joint tissue samples. d) Mankin and Osteoarthritis Research Society International (OARSI) scores of the mouse knee joint. n = 6. Data were analyzed using one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test. **p < 0.01 compared to Sham mice; ##p < 0.01 compared to OA mice; &&p < 0.01 compared to OA + MSC^IPFP Exo mice. Exo, exosome; MSC^IPFP, infrapatellar fat pad mesenchymal stem cells.

Wogonin-MSC^IPFP-Exos inhibited chondrocyte ECM degradation

According to immunohistochemical (IHC) staining of collagen II and MMP13 levels within the knee joints of mice, collagen II was notably downregulated and MMP13 was markedly upregulated within the knee joint tissue samples of OA mice, MSC^IPFP-Exos notably reversed the above effects of DMM modelling, and the effect of Wogonin-MSC^IPFP-Exos was more significant (Figures 6a to 6c). Western blot revealed that Sox9, aggrecan, and collagen II levels were markedly reduced, whereas ADAMTS5 and MMP13 levels were considerably increased within cartilage tissue samples of OA mice. MSC^IPFP-Exos and Wogonin-MSC^IPFP-Exos distinctly reversed the above impacts of the OA model, and Wogonin-MSC^IPFP-Exos exhibited a better ECM remodelling effect (Figure 6d). Together, these findings uncover that Wogonin-MSC^IPFP-Exo inhibited ECM degradation of chondrocytes, thereby alleviating degenerative joint lesions in OA mice.

Fig. 6.

Wogonin-MSC^IPFP-Exos inhibited chondrocyte extracellular matrix (ECM) degradation. a) to c) Immunohistochemical (IHC) staining of type II collagen (collagen II) and matrix metalloproteinase 13 (MMP13) within cartilage tissues of tibial plateau of knee joints (n = 4 for each group). d) Using Western blot, Sox9, proteoglycan (aggrecan), collagen II, a disintegrin and metalloproteinase with thrombospondin motifs 5 (ADAMTS5), and MMP13 levels within cartilage tissues of OA mice (n = 3 for each group) were evaluated. n = 3. Data were analyzed using one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test. **p < 0.01 compared to Sham mice; #p < 0.05 and ##p < 0.01 compared to OA mice; &p < 0.05 and && p < 0.01 compared to OA + MSC^IPFP Exo mice. Exo, exosome; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; MSC^IPFP, infrapatellar fat pad mesenchymal stem cells.

To explore the mechanisms underlying the effects of Wogonin-MSC^IPFP-Exo, we examined the expression of genes associated with Exo synthesis and release in MSC^IPFP in response to wogonin treatment, including SNAP23, Rab27a, Rab7a, Rab5, ATG5, and ATG16L1. Wogonin treatment significantly upregulated the gene expression levels of SNAP23, Rab5, ATG5, and ATG16L1 in MSC^IPFP (Figure 7b), suggesting enhanced Exo biogenesis and release. Next, we evaluated the expression of miRNAs implicated in chondrocyte activity within Wogonin-MSC^IPFP-Exo. Compared with MSC^IPFP-Exo, the levels of miR-145, miR-221, miR-140-5p, and miR-100-5p were increased in Wogonin-MSC^IPFP-Exo (Figure 7c). These miRNAs are known to be delivered by Exo to regulate chondrocyte proliferation, differentiation, or apoptosis.^24,29,30 To confirm the downstream effects of these miRNA changes, we analyzed wnt/β-catenin and mTOR signalling in chondrocytes treated with Wogonin-MSC^IPFP-Exo. Western blot analysis revealed that Wogonin-MSC^IPFP-Exo significantly reduced mTOR phosphorylation and increased β-catenin expression, indicating enhanced autophagy and anabolic signalling in chondrocytes (Figure 7d). Collectively, Wogonin treatment significantly enhanced Exo biogenesis in MSC^IPFP by upregulating key regulatory genes, including SNAP23, Rab5, ATG5, and ATG16L1, involved in Exo synthesis and release. Additionally, miRNA analysis revealed that Wogonin-MSC^IPFP-Exo contained elevated levels of miR-145, miR-221, miR-140-5p, and miR-100-5p. These Exo-delivered miRNAs are known to regulate chondrocyte proliferation, differentiation, and apoptosis via the wnt/β-catenin and mTOR pathways, which likely contributes to the enhanced therapeutic efficacy of Wogonin-MSC^IPFP-Exo (Figure 7a).

Fig. 7.

Wogonin pretreatment enhances MSC^IPFP exosome (Exo) biogenesis and modulates microRNA (miRNA) cargo to regulate chondrocyte signalling. a) Schematic illustration of the effects of wogonin pretreatment on MSC^IPFP-derived Exo biogenesis and downstream signalling in chondrocytes. b) Quantification of Exo biogenesis-related gene expression (Rab27a, Rab7a, SNAP23, Rab5, ATG5, and ATG16L1) in MSC^IPFP with or without wogonin pretreatment, measured by quantitative real-time polymerase chain reaction (qRT-PCR). c) Expression levels of miR-100-5p, miR-221, miR-140-5p, and miR-145 in MSC^IPFP-Exo and Wogonin-MSC^IPFP-Exo, measured by qRT-PCR. d) Western blot analysis of β-catenin and p-mTOR/mTOR levels in chondrocytes treated with MSC^IPFP-Exo or Wogonin-MSC^IPFP-Exo. Data are presented as mean (SD), and were analyzed using independent-samples two-tailed Student’s t-test. **p < 0.01. GAPDH, glyceraldehyde-3-phosphate dehydrogenase; lncRNA, long non-coding RNA; miRNA, microRNA; mRNA, messenger RNA; MSC^IPFP, infrapatellar fat pad mesenchymal stem cells; mTOR, mammalian target of rapamycin; ns, non-significant.

Discussion

IPFP, also known as Hoffa’s fat pad, was first described by Albert Hoffa in 1904.³¹ It is located in the anterior knee compartment, between the joint capsule and synovium, and is histologically dominated by adipocytes.³² In the past, the main role of IPFP has been cushioning as well as lubrication; however, as reported recently, IPFP might contribute to OA development via interactions with synovium, articular cartilage, and subchondral bone.^24,33 Ideally, the cells used to repair cartilage defects should have strong chondrogenic differentiation ability, strong cell proliferation ability, and be able to maintain their stemness during the repair process. In this context, MSC^IPFP has been shown to possess superior chondrogenic potential compared to other MSC sources due to their intra-articular location. Beyond this, our study provides novel insights by demonstrating that wogonin pretreatment significantly enhances the therapeutic efficacy of MSC^IPFP-derived Exos. Specifically, Wogonin-MSC^IPFP-Exo not only amplified the upregulation of key cartilage matrix components such as Sox9, aggrecan, and collagen II, but also more effectively suppressed the catabolic enzymes MMP13 and ADAMTS5, compared to standard MSC^IPFP-Exo. These findings suggest that wogonin pretreatment modifies the cargo composition of Exos, potentially through mechanisms involving miRNA and protein modulation, thereby enhancing their capacity to promote chondroprotection and cartilage repair. This distinct enhancement underscores the value of optimizing Exo-based therapies through pharmacological pretreatment, providing a new avenue for advancing OA treatment.

When OA occurs, chondrocytes are induced by external stimuli to synthesize cartilage matrix catabolic metabolizing enzymes, such as MMP13 and ADAMTS5, and to reduce collagen II and aggrecan, thereby destroying the ECM of cartilage.^28,34,35 In the meantime, Sox9 is an essential transcription factor contributing to OA development,²⁷ and Sox9 exerts a crucial effect on articular cartilage formation and self-repair.³⁶ Reportedly, Sox9 could drive aggrecan and collagen II expression levels.^37,38 Herein, we observed that IL-1β and DMM were able to induce downregulation of Sox9, aggrecan, and collagen II expression in chondrocytes and cartilage tissues, and increases in the expression levels of MMP13 and ADAMTS5. Previous studies on MSC-derived Exos have highlighted their potential to regulate chondrocyte activity and repair cartilage damage by modulating ECM components and inflammatory responses.^39-41 For instance, MSC Exos have been reported to promote Sox9, collagen II, and aggrecan expression while inhibiting MMP13 and ADAMTS5,^42,43 effects that were observed in our study as well. Notably, wogonin pretreatment appears to enhance these effects, possibly by influencing the composition of exosomal cargo. This aligns with findings that the functional efficacy of MSC Exos is highly dependent on their cargo, which can be modulated by environmental or pharmacological factors.⁴⁴ By demonstrating that wogonin pretreatment further amplifies these beneficial effects, our study provides additional support for the therapeutic relevance of MSC Exos in OA, and highlights the importance of optimizing their cargo composition for improved efficacy.

Wogonin has been shown to play a therapeutic role in several diseases,⁴⁵ primarily due to its anti-inflammatory and antioxidant properties. In this study, wogonin enhances Exo biogenesis and release by upregulating genes related to Exo synthesis in MSC^IPFP, including SNAP23, Rab27a, Rab7a, ATG5, and ATG16L1.⁴⁶ Furthermore, miRNA analysis revealed that Wogonin-MSC^IPFP-Exo exhibited increased levels of miR-145, miR-221, miR-140-5p, and miR-100-5p. These miRNAs have been reported to regulate chondrocyte activity.^24,47-49 Reportedly, miRNAs are involved in OA development,⁵⁰ whereas wogonin has been reported to modulate miR-145, miR-27b-5p, and miR-155 in vascular smooth muscle cells and cancer cells.^51-53 Therefore, the changes of the miRNA may be the potential mechanism. In addition, Wogonin-MSC^IPFP-Exo significantly downregulated mTOR activity while upregulating β-catenin expression in chondrocytes, which is critical for chondrocyte autophagy and chondrocyte differentiation.^54,55 This novel insight into the regulatory mechanisms of Wogonin-MSC^IPFP-Exo provides a mechanistic basis for their enhanced therapeutic efficacy in OA, suggesting a potential strategy for optimizing Exo-based therapies. Compared to other established strategies for enhancing MSC Exo efficacy, such as hypoxic preconditioning and treatment with growth factors, wogonin pretreatment offers unique advantages. Hypoxic preconditioning has been shown to increase exosomal yield and enhance the anti-inflammatory properties of MSC Exos, while growth factor stimulation often upregulates key signalling molecules beneficial for tissue repair. For example, wogonin significantly ameliorates the altered levels of oxidative stress markers, anti-oxidant proteins, and pro-inflammatory factors dose-dependently, which in turn markedly mitigates the symptoms in rheumatoid arthritis rat models.⁵⁶ In contrast, wogonin pretreatment (Wogonin-MSC^IPFP-Exos) not only augments Exo production but also specifically enhances the regulation of cartilage matrix components such as Sox9, aggrecan, and collagen II expression, while downregulating MMP13 and ADAMTS5. These effects suggest that wogonin pretreatment may target both inflammatory and degenerative pathways more comprehensively. However, further exploration of the therapeutic mechanism of Wogonin-MSC^IPFP-Exo in the inflammatory aspect is needed. Extensively, studies reported that MSC-isolated Exos cargo non-coding RNAs could be delivered to chondrocytes to play therapeutic benefits in OA via modulating targeted gene expression.⁵⁷ Despite these promising results, challenges remain for clinical translation. The scalability of wogonin-pretreated MSC-derived Exo production must be addressed, particularly for large-scale manufacturing under good manufacturing practices (GMP) conditions.⁵⁸ Regulatory hurdles, including the need for extensive preclinical safety and efficacy studies, must also be navigated. Moreover, the cost-effectiveness of wogonin pretreatment compared to other strategies is a critical factor for its adoption in clinical practice. Addressing these challenges will be essential to realize the full potential of wogonin-pretreated MSC Exos as a therapeutic option for OA.

In summary, this present study investigates the function of wogonin-MSC^IPFP-Exos in OA, which could promote cartilage ECM repair by upregulating the expression levels of the key transcription factor for cartilage repair, Sox9, and its target genes, aggrecan, and collagen II, and downregulating the cartilage matrix catabolic enzymes, MMP13 and ADAMTS5, thereby alleviating the damage of chondrocytes and cartilage tissue at both in vitro and in vivo OA model. This study continued to enrich and complement the mechanism of wogonin in the treatment of OA and provided a novel rationale for the use of wogonin in the treatment of OA. However, there are some limitations. First, we observed that wogonin pretreatment could enhance the function of MSC^IPFP-Exos for OA therapy, but the specific mechanisms responsible for this effect were not fully illustrated. In particular, the potential alterations in the exosomal cargo, such as miRNAs or proteins, induced by wogonin pretreatment have not been fully elucidated. This represents a key area for future research, as profiling the proteome and miRNA content of wogonin-pretreated Exos could provide valuable insights into their enhanced therapeutic effects. Second, in vivo evaluation of safety and duration of drug action of wogonin-MSC^IPFP-Exos is lacking. Given the promising results regarding the efficacy of wogonin-MSC^IPFP-Exo in promoting chondrocyte proliferation and suppressing apoptosis, further preclinical and clinical studies are necessary to validate the safety, efficacy, and long-term outcomes of this therapeutic approach before its translation into clinical practice for the treatment of OA. Strategies to extend the retention time of Exos and reduce the injection frequency are crucial for future clinical applications. To address this, biomaterial-based delivery systems, such as hydrogels or nanoparticle carriers, which have been reported to improve the stability and retention of Exos at the target site, are planned to be explored in future studies. Incorporating these approaches could enhance the therapeutic potential of Exos while minimizing the required injection frequency, thereby improving clinical translation and patient compliance. Determining the most suitable stem cell type for cartilage injury repair involves several challenges, including ensuring strong chondrogenic differentiation ability, high proliferation rates, and maintenance of stemness. The tissue of origin plays a crucial role in influencing MSC differentiation ability. MSCs derived from the infrapatellar fat pad have demonstrated significant chondrogenic differentiation potential, likely due to their intraarticular location and close proximity to articular cartilage,^59,60 making them a promising candidate for cartilage repair applications. In addition, chondrocytes were isolated from both severely and slightly damaged regions of the medial tibial cartilage. This heterogeneity might have influenced the phenotype of the isolated chondrocytes, potentially introducing variability in their response to experimental treatments. Future studies could consider isolating chondrocytes from more uniform regions to reduce variability and better define their phenotype.

Data Availability

All data generated or analyzed during this study are included in the published article and/or in the supplementary material.