Exosomes derived from berberine-treated bone marrow mesenchymal stem cells ameliorate inflammatory arthritis in rats with collagen-induced rheumatoid arthritis

ABSTRACT To evaluate the effect of exosomes secreted by berberine-treated bone marrow-derived mesenchymal stem cells (BMSCs) on synovial inflammation in collagen-induced arthritis (CIA) model. Exosomes were isolated from berberine-treated BMSCs (Ber-BMSC-EXs) and untreated BMSCs (BMSC-EXs). CIA rats were developed and treated with Ber-BMSC-EXs or BMSC-EXs. Clinical arthritis index was measured, activation of fibroblast-like synoviocytes (FLSs) were evaluated, and response and frequency of T helper 17 (Th17) and regulatory T (Treg) cells as well as cytokine levels in synovial fluid were determined. Ber-BMSC-EXs (with a superior effect than BMSC-EXs) significantly alleviated RA clinical score and paw inflammation, balanced Th17 and Treg cells’ responses and population, modulated levels of pro-inflammatory and anti-inflammatory cytokines in joint synovial fluid, and inhibited FLS activation in CIA rats. Ber-BMSC-EXs (with a significantly greater effect than BMSC-EXs) could significantly ameliorate synovial joint inflammation and clinical symptoms of arthritis in CIA rats, showing that berberine can reinforce anti-arthritic effects of BMSC-EXs.


Introduction
Rheumatoid arthritis (RA) is a chronic progressive inflammatory disease that predominantly attacks the synovial tissues and joints.RA is mainly characterized by chronic synovial inflammation (synovitis) and hyperplastic synovial cells (synovial hyperplasia), caused by inflammatory cell infiltration in the synovium, which leads to progressive and irreversible erosion of the bone and cartilage.Indeed, the inflammatory milieu in the synovial cavity provokes a vigorous tissue response, mainly activation of fibroblastlike synoviocytes (FLS) with a tumour-like inflammatory proliferation and invasive phenotype, which contribute to synovial inflammation and hyperplasia.Activated FLSs secrete numerous inflammatory mediators and proteases, which lead to the formation of synovial pannus and the penetration of the inflamed hyperplastic synovium into the neighbour articular structures, causing cartridge and bone damage.Pro-inflammatory and tissue-damaging cellular responses in synovitis are integrated by cytokines, such as tumour necrosis factor (TNF)-α, interleukin (IL)-1β, IL-6, and IL-17, which are produced by activated FLSs and various infiltrated immune cells (Bustamante et al., 2017;Tu et al., 2018).Among immune cells, CD4 + T cells including T helper 17 (Th17) and regulatory T (Treg) cells have the key roles in the RA pathogenesis.Th17 cells are pro-inflammatory cells producing IL-17 cytokine that exacerbates the synovial inflammation and the joint destruction, while Treg cells, which are weakly differentiated during RA, are anti-inflammatory cells producing IL-10 and transforming growth factor beta (TGF-β) cytokines that suppress Th17-mediated responses by reducing differentiation and homing of Th17 cells at the inflammatory joints (Banerjee et al., 2021;Baroroh et al., 2021;Shetty et al., 2021;You et al., 2021;Zhai et al., 2019).
The inflammation inhibition is the main mechanism underlying the preventive effects of the available RA therapeutics on the joint damage and bone loss.Currently, prescribed therapeutics used for the RA treatment, such as steroidal and non-steroidal anti-inflammatory drugs, disease-modifying anti-rheumatic drugs, and biologic agents, often provide relatively limited efficacy together with a momentary relief, and result in many severe side effects with long-term use (Crossfield et al., 2021;Leman et al., 2021).It is thus needed to explore alternative anti-arthritic medications with higher effectiveness and reliability.
Exosomes have been known as effectual players in the RA progression (Tavasolian et al., 2020).Exosomes are nano-sized (below 150 nm) extracellular vesicles that are actively secreted by various types of cells and act as mediators in cell-to-cell communications by transferring functional cargoes, such as proteins, lipids, and genetic materials, thereby affecting the behaviour of recipient cells by modulating biological processes and cellular signallings (Tavasolian et al., 2020;Zhai et al., 2022).Mesenchymal stem cells (MSCs), such as bone marrow-derived mesenchymal stem cells (BMSCs), have been found to release a mass of exosomes with the immunomodulatory-mediated therapeutic activity (Burrello et al., 2016;Cosenza et al., 2017;Yeo et al., 2013).Of note, recent studies showed the interest of using BMSC-derived exosomes (BMSC-EXs) to alleviate clinical symptoms of RA.An early study indicated that BMSC-EXs exert the anti-inflammatory effect on B and T lymphocytes in experimental models of inflammatory arthritis (Cosenza et al., 2018).Subsequent studies have shown that BMSC-EXs can inactivate RA-related FLSs (RA-FLSs), whereby reducing clinical symptoms of RA in collagen-induced arthritis (CIA) inflammatory model (Chen et al., 2018;Meng et al., 2020;Meng & Qiu, 2020;Su et al., 2021).
With reference to these findings and also similar effects of berberine and BMSC-EXs on RA in the CIA model, we encouraged to determine whether berberine affects antiarthritic effect of BMSC-EXs in CIA rats.To this end, we isolated exosomes secreted by BMSCs treated with and without berberine and evaluated their effects on in vitro inflammatory proliferation of FLSs isolated from CIA rats (CIA-FLSs) and also on clinical symptoms and joint inflammation in CIA rats.

Cell culture
Rat BM-MSCs were obtained from the Cell Bank of Chinese Academy of Sciences (Shanghai, China) received with the fifth passage and utilized within the seventh to the eighth passages in accordance with the manufacturer's protocols.The cells were cultured in low-sugar Dulbecco's modified Eagle's medium/F12 (DMEM; Gibco) containing 10% exosome-free fetal bovine serum (FBS; Gibco), 2 mM L-glutamine (Sigma), 100 mg/ ml streptomycin and 100 U/ml penicillin (Gibco), at 37°C in an incubator with 95% humidity and 5% CO 2 .

Exosome isolation
BM-MSCs (2.5 × 10 6 cells) were seeded in a T175 flask and the culture supernatant was collected after 48 h of treatment with or without berberine (10 μM).Differential centrifugation and ultracentrifugation approach was used to isolate exosomes released by treated or untreated cells, as previously reported (Li, Stöckl, et al., 2021;Osterman et al., 2015;Qiu et al., 2020;Wu, Zhou, et al., 2016).Briefly, the cell debris/dead cells and other contaminations in the cell culture supernatant were eliminated by sequential steps of centrifugation; 10 min at 300g, 15 min at 2500g, followed by 25 min at 12,000g, all at 4°C.The supernatant were then ultra-centrifuged for 80 min at 110,000g to separate exosomes, which were then dispersed in the Phosphate-buffered saline (PBS) buffer, filtered through a sterile 22-μm filter, and ultra-centrifuged again for another 80 min at 110,000g.Finally, the pellet was resuspended in the PBS buffer containing 25 mM Trehalose and stored at −80°C.To measure the concentration of isolated exosomes, a bicinchoninic acid assay (BCA) protein assay kit was employed in accordance with the manufacturer's protocols (Abcam, Cambridge, UK).

Dynamic light scattering
The size distribution of vesicles in exosome isolates (Ber-BMSC-EXs and BMSC-EXs) was determined using dynamic light scattering (DLS) by a Zetasizer (Malvern Instruments, Malvern, UK).Exosome isolates were diluted in the PBS buffer (1:200) and measurements were recorded for 30 s at 25°C.Three replicates of each sample were measured and averaged to present the size and concentration values.

Transmission electron microscopy
The size and morphology of vesicles in exosome isolates (BMSC-EXs and Ber-BMSC-EXs) were monitored by transmission electron microscopy (TEM).Vesicles were decorated by negative staining to be visualized by a TEM instrument (Philips CM10 TEM; ×200,000 magnification).In brief, freshly isolated exosomes were dispersed in the cold PBS buffer and loaded onto parafilm.The loaded fluid was covered using a formvar (polyvinyl formal)-carbon-coated 400 copper mesh grid for 10 min at 25°C, followed by 1 min of incubation with 2% phosphotungstic acid and drying at 25°C for 15 min.

Animals
Male Sprague-Dawley rats (5-7 weeks old, 160-180 g) were obtained from Experimental Animal Centre of Shandong University.Animals were kept in the pathogen-free condition with the controlled temperature (24 ± 2°C) and humidity (55 ± 5%), under a 12 h dark/light cycle, with ad libitum access to tap water and standard chow diet.The priority was to decrease the number of animals used and to minimize animals' suffering.

Development and assessment of collagen-induced RA (CIA) in rat
CIA was developed in rats according to the previously published method (Brand et al., 2007;Rosloniec et al., 2010), with some modifications.In brief, type II collagen (CII; 4 mg/ml; Sigma) was emulsified in complete Freund's adjuvant (CFA; 4 mg/ml; Sigma) at 1:1 ratio.On day zero, rats were immunized with 200 µl CII/CFA emulsion via a slow subcutaneous injection at 1.5 cm distal from the base of the tail, followed by a booster immunization on day 7 with 100 µl emulsion at a tail area proximal to the primary injection site.Rats were daily checked for signs and symptoms of arthritis by two blinded, expert examiner.The arthritis severity was determined by scoring the inflamed paw according to the clinical arthritis index graded from 0 to 4, in which 0 = no swelling or erythema, 1 = swelling or erythema of one digit, 2 = swelling or erythema of two digits, 3 = swelling or erythema of more than two digits and footpad, and 4 = swelling or erythema of the entire paw and joint rigidity with incapacity to bend the knee (Choi et al., 2016;Tong et al., 2015).The total score of four paws with a maximum score of 16 for each rat was employed to determine the overall severity and progression of RA.The incidence of disease was denoted as the appearance of inflamed paws with a score ≥2 among four paws (Choi et al., 2016).The swelling volume was determined in paws with a foot volume plethysmometer once a week after CII/CFA immunization.

Treatment and grouping
The primary signs of arthritis in CIA rats were observed in knee joints after 4 weeks.Rats were allocated into four experimental groups (n = 8) on day 28, including the normal control group, the CIA model group, CIA + BMSC-EXs group, and CIA + Ber-BMSC-EXs group.In the two latter groups, the knee joints of CIA rats were injected intra-articularly with BMSC-EXs (100 µg/100 µl) and Ber-BMSC-EXs (100 µg/100 µl), respectively, on a weekly basis for 4 weeks from day 28, under anaesthesia.In the two former groups, normal rats and CIA rats received intra-articular injection of PBS buffer (100 µl) in the same schedule.Of note, the treatment dose, administration route, and power of study were selected in accordance with prior reports that demonstrated beneficial in vivo effects in rodents (Qi et al., 2016;Zhang et al., 2016).
One week after the last treatment, all rats were euthanized using an extra dosage of chloral hydrate hydrochloride, and subsequent experimental studies on the knee joints and blood samples were conducted as explained in the following sections.

Isolation and culture of CIA-FLSs
CIA-FLSs were isolated by enzymatic digestion method from synovial tissues of CIA rats with reference to an already published protocol (Zhao et al., 2016).On day 28 after CIA and under sterile conditions, synovial tissues were dissected form the knee joints, cut into small sections (<1 mm), washed gently in the sterile PBS buffer, and digested through incubation in the PBS containing type I collagenase (1 mg/ml) at 37°C for 2 h.The obtained cell suspension was incubated overnight in low-sugar DMEM medium containing 10% FBS, 2 mM L-glutamine, 100 mg/ml streptomycin, and 100 U/ml penicillin, in 95% air humidity with 5% CO 2 at 37°C.Afterward, non-adherent cells and cell debris were eliminated and a homogenous population of adherent CIA-FLSs was cultured under a similar condition for subsequent experiments.

Purity assessment of CIA-FLSs
The purity of cultured CIA-FLSs (passage 3) was evaluated using flow cytometry analysis of the two fibroblast markers, CD90.1 (Thy-1) and CD54 Intercellular adhesion molecule (ICAM-1), in accordance with manufacture's instructions (Biolegend).Briefly, 1 × 10 5 cells were stained with phycoerythrin (PE)-tagged anti-rat CD90.2 and anti-rat CD54 monoclonal antibodies (Biolegend), through incubation in the stain buffer (PBS containing 1% bovine serum albumin (BSA)) at 4°C for 40 min.PE-tagged mouse IgG1 isotype antibody (Biolegend) was also used as the negative control.Subsequently, a flow cytometry instrument (FACS Calibur; BD Biosciences, San Jose, USA) was employed to evaluate cell fluorescence, and FlowJo software was then used to analyse the obtained data.

Cell viability/proliferation assay
To evaluate the effect of Ber-BMSC-EXs and BMSC-EXs on the proliferation of isolated CIA-FLSs, the cell viability assay was carried out by the MTT [3-(4,5-dimethylthiazol-2yl)-2,5-diphenyl-tetrazolium bromide] assay (Sigma-Aldrich).In brief, cells were seeded in the 96-well plates at a density of 1 × 10 4 cells/well, and incubated in 5% CO 2 at 37°C for 24 h.Thereafter, the cultured cells were treated with Ber-BMSC-EXs (100 µg/ml) or BMSC-EXs (100 µg/ml) or vehicle control (PBS), and incubated for 24, 48, and 72 h.Each treatment was repeated six times.The cells were then incubated with MTT solution (5 mg/ml) for 4 h.Afterward, the produced formazan crystals were dissolved in dimethyl sulfoxide, and absorbance at 560 nm was detected using an ELISA plate reader (Bio-Tek).The results were presented in terms of cell viability % in comparison with control cells as the mean ± standard deviation (SD) of three independent assays.

Assessment of synovial inflammation
To determine the effect of Ber-BMSC-EXs and BMSC-EXs on the synovial inflammation in CIA rats, we evaluated Th17 and Treg cells (using flow cytometry analysis) as well as cytokine levels (using ELISA assay) in the synovial fluids of knee joints at the end of study (week 8) and cytokine production by isolated CIA-FLSs as explained in the following subsections.

Collection of synovial fluid
To collect the synovial fluid of the knee joint, the already reported protocols were used with some modifications as follows.After sacrificing rats, the skin and the patellar ligaments were excised to uncover the synovial membrane of the knee.The needle (30gauge) of a 0.3 ml syringe filled with 250 μl of heparinized PBS was inserted into the membrane, and the articular cavity was washed via slow and repetitive infusions and aspirations with PBS to collect synovial lavage ingredients.The process was twicely repeated to collect 500 μl of synovial fluid sample.

Isolation of synovial fluid mononuclear cells
To evaluate the presence of Th17 and Treg cells in knee joints, mononuclear cells were isolated from synovial fluids in accordance with previously published protocol, as follows.Freshly isolated synovial fluids were centrifuged at 700g for 8 min at 23°C (supernatant was collected for the cytokine assay by the ELISA).Cell pellets were resuspended in Roswell park memorial institute (RPMI) 1640 supplemented with density gradient cell separation solution (Sigma) and were centrifuged at 1500g for 30 min at 22°C.The cells were then washed by PBS buffer through centrifugation (repeated three times) at 400g for 8 min at 23°C.The isolated cells were then resuspended in the RPMI 1640 medium, and live cells were counted by the trypan blue-staining and analysed by flow cytometry.

Flow cytometry analysis
The presence of Th17 cells (CD4 + IL-17 + T cells) and Treg cells (CD4 + CD25 + Foxp3 + T cells) in synovial fluids was determined using flow cytometry analysis.In brief, the isolated synovial fluid mononuclear cells were incubated for 4 h at 37°C in the complete RPMI 1640 medium containing phorbol myristate acetate (50 ng/ml), ionomycin (500 ng/ml), and Brefeldin A (5 μg/ml).Afterward, cells were stained by fluorescenttagged monoclonal antibodies against surface markers, including fluorescein isothiocyanate-tagged anti CD4, PE-tagged anti CD25 (BD Biosciences).For intracellular staining, cells were first fixed and permeabilized by the Cytofix/Cytoperm kit (BD Biosciences) according to manufacturer's instructions, then stained with PE-tagged anti IL-17 and PE-Cy5-tagged anti FoxP3 antibodies (BD Biosciences).Finally, stained cells were washed with the wash buffer (BD Biosciences), suspended in stain buffer (BD Biosciences), and analysed by a FACS-Calibur flow cytometer (BD Biosciences, San Jose, USA).

ELISA
To assay cytokines in the knee joint, the supernatants obtained after centrifugation of synovial fluid samples were used.The levels of pro-inflammatory cytokines including TNF-α, IL-1β, IL-6, and IL-17 as well as anti-inflammatory cytokines including IL-10 and TGF-β in synovial fluids were assayed using specific ELISA kits (MyBioSource or R&D Systems) in accordance with the manufacturer's protocols.To evaluate cytokine production by the CIA-FLSs, cells (5 × 10 4 cells/well) were seeded in a 12-well plate, followed by 24-h incubation in 5% CO 2 at 37°C, and then treated with Ber-BMSC-EXs (100 µg/ml) or BMSC-EXs (100 µg/ml).After 24-h of treatment, the levels of TNF-α, IL-1β, and IL-6 were assayed in the culture medium using ELISA in accordance with the instructions of the kits.

Statistical analysis
The results were analysed using GraphPad Prism software (Version 9).The values were presented as mean ± SD or median [interquartile range] for normally and non-normally distributed data, respectively.The statistical analysis was carried out using a one-way analysis of variance followed by Tukey's post hoc test to assess the significance of the differences among the experimental groups.Data with p-value < 0.05 show the statistical significance.

Exosome identification
The result of the acetylcholinesterase activity assay demonstrated a significantly high enzyme activity in samples isolated from berberine-treated and untreated BMSCs, whereas it was non-detectable in the control sample, verifying the presence of exosome particles in the isolates (Figure 1(A)).It was further verified via evaluation of the size and the morphology of the particles in the isolates using DLS and TEM analysis.The particles in both samples isolated from berberine-treated and untreated BMSCs had a circular disk-like shape within the size range of exosomes, with mean diameters around 80-110 nm (Figure 1(B-D)).

Purity of isolated CIA-FLSs
The flow cytometry analysis of isolated synovial cells indicated the presence of a high percentage of cells positive for fibroblast markers CD90.2 (91.7%) or CD54 (88.3%) (Figure 2).The results confirmed the isolation of FLSs with a high purity from knee joints of CIA rats.

Ber-BMSC-EXs inhibit the activated CIA-FLSs
The impact of Ber-BMSC-EXs on the proliferation of CIA-FLSs was assessed using the MTT colorimetric assay.The isolated CIA-FLSs (at passage 3) were treated with Ber-BMSC-EXs, BMSC-EXs, or the vehicle control for different durations (24, 48, and 72 h).The results indicated that there was a significant difference (p < 0.001) between the proliferation of CIA-FLSs in exosome groups and control groups (Figure 3(A)).Of note, both Ber-BMSC-EXs and BMSC-EXs significantly (p < 0.001) decreased the percentage of viable CIA-FLSs, when compared to the vehicle control, in a time-dependent manner (Figure 3(A)).However, a significantly higher decrease (p < 0.001 or p = 0.005) in the percentage of viable cells was found in Ber-BMSC-EXs group than BMSC-EXs group during all treatment time points (Figure 3 (A)).Furthermore, the production of pro-inflammatory cytokines by CIA-FLSs treated with the Ber-BMSC-EXs or the BMSC-EXs were also measured.It was found that treatment with Ber-BMSC-EXs or BMSC-EXs could significantly decreased the production of pro-inflammatory cytokines including TNF-α, IL-6, and IL-1β by CIA-FLSs, when compared with the control treatment (Figure 3(B-D)).Significantly, the modulatory effect of Ber-BMSC-EXs on the cytokine production was found to be more profound than that of BMSC-EXs.

Ber-BMSC-EXs inhibit the progression of inflammatory arteritis in CIA rats
To investigate the anti-RA impacts of Ber-BMSC-EXs, we employed the CIA rat model administrated with four intra-articular injections of Ber-BMSC-EXs (100 µg/rat) in weekly intervals from week 4 to 7 after the first immunization.The progression of RA was monitored until week 8. Clinical symptoms of RA were assessed in the paws.Ber-BMSC-EXs treatment was started after the initiation of severe CIA, 4 weeks after the first immunization.In CIA rats, the clinical scores of RA (Figure 4(A)) and the degrees of paw swelling (Figure 4(B)) were significantly elevated compared to control rats.However, 4 weeks treatment with weakly administration of Ber-BMSC-EXs or BMSC-EXs significantly reduces the RA clinical score (Figure 4(A)) and paw swelling volume (Figure 4(B)) in CFA rats.Notably, these were significantly decreased in CIA + Ber-BMSC-EXs rats when compared to CIA + BMSC-EXs rats (Figure 4(A,B)).

Ber-BMSC-EXs modulate the levels of cytokines in the joint synovial fluid
A characteristic feature of CIA is a remarkable increase in the expression of pro-inflammatory cytokines.At the end of study, the levels of pro-inflammatory cytokines and antiinflammatory cytokines in the joint synovial fluid were assayed by ELISA.The levels of pro-inflammatory cytokines, including IL-1β (p < 0.01), IL-6 (p < 0.01), IL-17 (p < 0.001), and TNF-α (p < 0.001), were significantly decreased in the CIA + Ber- BMSC-EXs group and CIA + BMSC-EXs group, when compared with the CIA group (Figure 5(A-D)).Notably, the levels of pro-inflammatory cytokines in the CIA + Ber-BMSC-EXs group were significantly lower than those in the CIA + BMSC-EXs group (p < 0.001) (Figure 5(A-D)).In case of anti-inflammatory cytokines, the levels of IL-10 and TGF-β were significantly higher in the CIA + Ber-BMSC-EXs group and the CIA + BMSC-EXs group compared to the CIA group (Figure 5(E,F)).However, the levels of the anti-inflammatory cytokines in the CIA + Ber-BMSC-EXs group were significantly higher than those in the CIA + BMSC-EXs group (Figure 5(E,F)).

Ber-BMSC-EXs modulate the Th17/Treg cells balance
The Th-17/Treg imbalance is a well-known impaired immune response in the RA pathogenesis; Th17 cells have a high frequency and exert a destructive role, whereas Treg cells have a protective role and an impaired function, leading to the joint inflammation and destruction.To evaluate the effect of intra-articular administration of exosome formulations on the cell responses and populations of Treg and Th17 cells in the joint synovial fluid of the CIA rats, ELISA and flow cytometry analysis were employed.Ber-BMSC-EXs or BMSC-EXs treatments significantly increased the proportion of CD4 + CD25 + Foxp3 + Treg cells (Figure 6) and the level of IL-10 and TGF-β, as markers of Treg response (Figure 5(E,F)), in the synovial fluid of CIA rats compared to the control treatment.On the other hand, the population of IL-7-producing CD4 + (Figure 7) and the level of IL-17 (Figure 5) were significantly decreased in the synovial fluid of CIA + Ber-BMSC-EXs or CIA + BMSC-EXs groups, when compared with the control group.Of note, the Ber-BMSC-EXs treatment showed a higher influence on the frequency and responses of Treg and Th17 cells in the CIA rats than the BMSC-EXs treatment.The inhibitory effect of Ber-BMSC-EXs on the RA progression in CFA rats.Ber-BMSC-EXs significantly decreased score of the arthritis index (A) and the volume of the paw swelling (B) in CIA rats compared with control rats.Of note, the inhibitory effects of Ber-BMSC-EXs on the activated CIA-FLSs were significantly higher than that of BMSC-EXs.The changes of arthritis scores (A) and the paw swelling (B) were evaluated at weekly intervals from week 4 to week 8 after the first immunization in different experimental groups.The results of the paw swelling volume assessment are represented as the mean ± SD (n = 8).The results of the arthritis index are expressed as the median [25% percentile − 75% percentile] (n = 8).*p = 0.02 statistical difference between CIA group and CIA + Ber-EXs group, $ p < 0.001 statistical difference between CIA group and CIA + Ber-EXs group, ***p < 0.001 statistical difference between CIA group and CIA + EXs group, and p < 0.001 statistical difference between CIA + EXs group and CIA + Ber-EXs, # p = 0.02 statistical difference between CIA + EXs group and CIA + Ber-EXs.
In the present study, we evaluated and compared for the first time the effect of the local intra-articular administration of exosomes derived from BMSCs (BMSC-EXs) and berberine-treated BMSCs (Ber-BMSC-EXs) on the in vitro proliferation of CIA-FLSs and on the inflammatory arthritis in the CIA model.The resultant data indicated that either BMSC-EXs or Ber-BMSC-EXs could significantly ameliorate the synovial joint inflammation and the clinical symptoms of arthritis in CIA rats.However, Ber-BMSC-EXs exerted a significantly greater effect than BMSC-EXs, showing that berberine can reinforce anti-arthritic effects of BMSC-EXs.
FLSs play a key role in the RA/CIA pathogenesis.In RA/CIA, FLSs are abnormally activated and exhibit tumour-like proliferation and invasion, which contribute to synovial hyperplasia, causing the synovial pannus formation and cartridge degeneration (Bustamante et al., 2017;Tu et al., 2018).Thus, inhibiting excessive proliferation of activated FLSs may attenuate the disease severity and provide an effective approach in the RA treatment.Several independent experimental investigations suggested that berberine or BMSC-EXs (Meng et al., 2020;Meng & Qiu, 2020;Su et al., 2021;Zhang et al., 2022) could exert the anti-arthritic effect through proliferation inhibition and apoptosis induction of the activated arthritis-FLSs.Of note, we found that Ber-BMSC-EXs, with a higher effect than BMSC-EXs, reduced the in vitro proliferation of CIA-FLSs.Moreover, activated FLSs secrete a huge amount of pro-inflammatory cytokines, causing synovial inflammation and further intensifying joint destruction, and eventually worsening the arthritis progression (Bustamante et al., 2017;Tu et al., 2018).TNF-α provokes the synovial inflammation and joint destruction during RA via inducing the adhesion and migration of inflammatory cells and promoting the synthesis of other pro-inflammatory cytokines (Bertolini et al., 1986;Edrees et al., 2005;Matsuno et al., 2002).IL-1β plays a critical role in synovial inflammation and pannus formation.In association with other inflammatory cytokines such as TNF-α and IL-6, IL-1β exacerbates the progression of joint inflammation and concomitant cartilage and bone erosion in RA/CIA (Joosten et al., 1996(Joosten et al., , 1999;;Nikfar et al., 2018;van den Berg et al., 1994;Yang et al., 2019).Increased levels of pro-inflammatory cytokines such as TNF-α, IL-1β, and IL-6 have been found in the synovial fluids from RA patients and CIA models.It was previously reported that BMSC-EXs decreased the production of inflammatory cytokines TNF-α and IL-1β in mouse FLSs, and inhibited the RA progression in the CIA mice (Li, Fang, et al., 2021).Interestingly, we found that Ber-BMSC-EXs, with a superior effect than BMSC-EXs, decreased the production of inflammatory cytokines including TNF-α, IL-1β, and IL-6 by CIA-FLS cells, in vitro.Consistently, we detected a significant reduction in the levels of these cytokines in the synovial fluid of the knee joints in the CIA rats treated with Ber-BMSC-EXs or BMSC-EXs; however, Ber-BMSC-EXs showed a greater effect than BMSC-EXs.Furthermore, clinical observations showed that the paw inflammation, a typical clinical sign of inflammatory CIA characterized by swelling and/or erythema, was significantly decreased in CIA rats treated with Ber-BMSC-EXs or BMSC-EXs; however, a higher effect was found with Ber-BMSC-EXs than BMSC-EXs.
A previous study showed that BMSC-EXs can improve Th17/Treg balance and efficiently alleviate inflammation and arthritis signs in the CIA mice (Cosenza et al., 2018).Our results indicated that Ber-BMSC-EXs or BMSC-EXs can regulate the balance between Treg and Th17 cells.Ber-BMSC-EXs, with a higher effect than BMSC-EXs, were found to decrease the population of T17 cells (IL-17-producing CD4 + T cells) and increase the population of Treg cells (CD25 + Foxp3 + T cells) in the joint synovial fluid of the CIA rats.It was further supported by the results that showed a reduction of Th17-related cytokine (IL-17) and an increase of Treg-related cytokine (IL-10) in the synovial fluids.Thus, our results indicate that Ber-BMSC-EXs or BMSC-EXs can modulate the phenotype of the activated CD4 + T cells from IL-17-producing cells towards Foxp3 + Treg cells in the joint synovial fluid of CIA rats.These may be due to the modulatory effect of the exosome formulations on the cytokine levels in the synovial fluid.Of note, the presence of IL-17, IL-1β, and IL-6 in the synovial fluid can promote differentiation of the Th17 cells (Amadi-Obi et al., 2007;Wu, Yan, et al., 2016), while TGF-β can promote Treg cell differentiation (Beal et al., 2012;Chen & Konkel, 2015); notably, Ber-BMSC-EXs or BMSC-EXs increased the levels of TGF-β and decreased the levels of IL17, IL-1β, and IL-6 in the joint synovial fluid.
Importantly, the improving effects of berberine on anti-arthritic effects of BMSC-EXs can be due to anti-inflammatory properties of berberine.Of note, there have been a large number of investigations showing anti-inflammatory properties of berberine in various diseases like RA (Ehteshamfar et al., 2020;Haftcheshmeh et al., 2022;Mohammadian Haftcheshmeh & Momtazi-Borojeni, 2021;Shen et al., 2020).Thus, it can be suggested that berberine may alter cargo content of BMSC-EXs, such as miRs and proteins/ enzymes that are responsible for anti-inflammatory properties of these exosomes, thereby improving the anti-arthritic feature of BMSC-EXs.
To the best of our knowledge, there is no study reporting the therapeutic efficacy of Ber-BMSC-EXs in the experimental inflammatory arthritis and, thus, the results of present study endow a novel therapeutic opportunity for BMSC-EXS in the RA treatment.However, although the results of our study clearly indicate a potential future therapy by berberine-primed exosomes, a number of limitations need to be acknowledged in the future studies.The impact of berberine on the cargo content of BMSC-EXS as well as the underlying molecular mechanisms should be evaluated.Moreover, the histopathological alterations in knee joints and the synovial tissue should also be assessed.Furthermore, it is essential to do in vivo toxicity assessments to determine non-toxic dose of Ber-BMSC-EXS in RA.

Figure 1 .
Figure 1.Exosome identification.Acetylcholinesterase activity was markedly increased in medium samples isolated from berberine-treated BMSCs and untreated BMSCs, verifying the presence of exosome particles (A).The size distribution analysis indicated the presence of vesicles in the size range of exosomes (80-110 nm) (B).Representative photomicrographs indicate the morphology and size of Ber-BMSC-EXs and BMSC-EXs visualized by TEM (×200,000 magnification; scale bar, 200 nm) (C and D).No significant differences were detected in acetylcholinesterase activity and size distribution of vesicle between Ber-BMSC-EXs and BMSC-EXs.Results are presented as mean ± SD of three independent experiments.

Figure 3 .
Figure 3.The inhibitory effect of Ber-BMSC-EXs on the activated CIA-FLSs.Ber-BMSC-EXs reduced significantly the proliferation of CIA-FLSs during 72-h post-treatment compared with the control (A).Ber-BMSC-EXs decreased significantly the production of inflammatory cytokines, including IL-1β, IL-6, and TNF-α, in CIA-FLSs compared with the control (B-D).Of note, the inhibitory effects of Ber-BMSC-EXs on the activated CIA-FLSs were significantly higher than that of BMSC-EXs.Results are presented as mean ± SD of three independent experiments.CIA, collagen-induced rheumatoid arthritis; FLSs, fibroblastlike synoviocytes.

Figure 4 .
Figure 4.The inhibitory effect of Ber-BMSC-EXs on the RA progression in CFA rats.Ber-BMSC-EXs significantly decreased score of the arthritis index (A) and the volume of the paw swelling (B) in CIA rats compared with control rats.Of note, the inhibitory effects of Ber-BMSC-EXs on the activated CIA-FLSs were significantly higher than that of BMSC-EXs.The changes of arthritis scores (A) and the paw swelling (B) were evaluated at weekly intervals from week 4 to week 8 after the first immunization in different experimental groups.The results of the paw swelling volume assessment are represented as the mean ± SD (n = 8).The results of the arthritis index are expressed as the median [25% percentile − 75% percentile] (n = 8).*p = 0.02 statistical difference between CIA group and CIA + Ber-EXs group, $ p < 0.001 statistical difference between CIA group and CIA + Ber-EXs group, ***p < 0.001 statistical difference between CIA group and CIA + EXs group, and p < 0.001 statistical difference between CIA + EXs group and CIA + Ber-EXs, # p = 0.02 statistical difference between CIA + EXs group and CIA + Ber-EXs.

Figure 5 .
Figure5.Ber-BMSC-EXs modulate the abnormal levels of cytokines in the joint synovial fluid of CIA rats.The levels of pro-inflammatory cytokines, including IL-1β, IL-6, IL-17, and TNF-α, were significantly decreased in CIA + Ber-BMSC-EXs when compared with the control rats (A-D).However, the levels of anti-inflammatory cytokines, including IL-10 and TGF-β, were significantly decreased in CIA + Ber-BMSC-EXs compared with those in the control rats (D, E).The modulatory effect of Ber-BMSC-EXs on anti-and pro-inflammatory cytokines was more profound than that of BMSC-EXs.Results are presented as mean ± SD of three independent experiments.

Figure 6 .
Figure 6.Ber-BMSC-EXs increase the population of Treg cells in the joint synovial fluid of CIA rats.(A) The representative flow cytometry data of CD25 + Foxp3 + Treg cells.(B) Percentage of CD25 + Foxp3 + Treg cells in the synovial fluid.Ber-BMSC-EXs show the superior effect than BMSC-EXs.Results are presented as mean ± SD of three independent experiments.

Figure 7 .
Figure 7. Ber-BMSC-EXs decrease the population of Th17 cells in the joint synovial fluid of CIA rats.(A) The representative flow cytometry data of CD4 + IL-17 + T cells.(B) Percentage of CD4 + IL-17 + T cells in the synovial fluid.Ber-BMSC-EXs show the superior effect than BMSC-EXs.Results are presented as mean ± SD of three independent experiments.