DBA/1J mice (male, 6–8 week old) were obtained from Nanjing Biomedical Research Institute of Nanjing University (Nanjing, China). C57BL/6J mice (male, 6–8 week old) were purchased from Liaoning Changsheng biotechnology company (Shenyang, China). All experiments using mice were performed in accordance with protocols approved by the Institutional Animal Care and Use Committee (IACUC) at China Medical University.
Isolation, culturing, and characterization of human GMSC
Human tissue samples were obtained from discarded tissues of patients who had relatively healthy periodontium undergoing routine dental procedures and who provided informed consent in the Dental Division of the Third Affiliated Hospital at Sun Yat-sen University. This study was carried out in accordance with the recommendations of the ethical review committee of clinical research of the Third Affiliated Hospital of Sun Yat-sen University. Human GMSC were obtained by following the protocol described previously [10, 11]. The cells were cultured with complete growth medium [MEM alpha (Gibco, Grand Island, NY) supplemented with 10% fetal bovine serum (Gibco, Grand Island, NY), 100 µg/mL penicillin and 100 µg/mL streptomycin (Gibco, Carlsbad, CA), 100 µM MEM Non-Essential Amino Acids (Gibco, Grand Island, NY), 550 µM 2-ME (Sigma-Aldrich, Darmstadt, Germany), 10 mM HEPES, 1 mM sodium pyruvate, 2 mM l-glutamine(Gibco, Paisley, UK)] at 37℃ in a humidified tissue culture incubator with 5% CO2.
We characterized GMSC by detecting the stem cell phenotypic markers and multipotent differentiation properties. Sub-cloning cultures were used to purify GMSC and cells from second to the third passages were used in experiments. For GMSC characterization markers detection, GMSC were stained with mAbs for human CD34, CD44, CD45, CD73, CD90, and CD105 (Biolegeng, San Diego, CA) and assessed by flow cytometry.
For GMSC multipotent differentiation properties detection, osteogenic differentiation and adipogenic differentiation were used. For osteogenic differentiation, the GMSC were seeded in 6-well plates (1×105 cells/well, Corning) and incubated with 2 ml of complete growth medium (α-DMEM). After GMSC reached 80–90% confluency, the cell medium was replaced with osteogenic induction medium: α-DMEM (Gibco, Grand Island, NY) containing 10% FBS, 0.1 µM dexamethasone (Sigma-Aldrich, Darmstadt, Germany), 10 mM β-glycerol phosphate (Sigma-Aldrich, Darmstadt, Germany), and 50 µg/ml L-ascorbic acid (Sigma-Aldrich, Darmstadt, Germany). The medium was refreshed every 3 days. Two weeks later, the cells were fixed and assayed by alizarin red staining kit (Sigma-Aldrich, Darmstadt, Germany).
For adipogenic differentiation, the GMSC were cultured in adipogenic differentiation medium: α-DMEM containing 500 nM isobutylmethylxanthine (Sigma-Aldrich, Darmstadt, Germany), 60 µM indomethacin (Sigma-Aldrich), 500 nM hydrocortisone (Sigma-Aldrich), 10 µg/ml insulin (Sigma-Aldrich), and 100 nM L-ascorbic acid phosphate. After 14 days, the cultured cells were stained with Oil Red-O (Sigma-Aldrich).
Production and characterization of GMSC-Exo
GMSC were seeded at 5×105 cells/ml on 10 cm dish with complete growth medium. When GMSC reached 80–90% confluence, they were cultured in conditioned medium with exosomes-free fetal bovine serum for 24 h, and then the supernatants containing GMSC-Exo were harvested. The GMSC-Exo were extracted by ultracentrifugation. In brief, the supernatants were centrifuged at 300 g for 10 min and 2,000 g for 10 min to remove the detached cells and debris/apoptotic bodies, respectively. Then, the supernatant was centrifuged for 30 min at 10, 000 g to remove macromolecular extracellular vesicles (EVs). After that, the clarified supernatant was centrifuged at 100,000 g for 70 min and the pellet on the bottom of the tube was GMSC-Exo. Then, the pellet was washed with phosphate- buffered saline (PBS) by centrifuging at 100,000 g for 70 min (Fig. 1A) . Finally, the concentrated GMSC-Exo were suspended in PBS and determined using a BCA protein assay kit (Beyotime, Nanjing, CHN).
Size distribution of GMSC-Exo was determined by Nanoparticle Tracking Analysis (NTA) using a Zetasizer Nano ZS90 instrument as advised by the manufacturer (Malvern Zetasizer Software v7.11 PSS0012-37 EN JP). Visualization of GMSC-Exo was assessed by Transmission Electron Microscopy (TEM, Hitachi HT7700, JP). GMSC-Exo suspensions were loaded on Formvar-coated grids and negatively stained with uranyl acetate for 15 min. Grids were observed using a microscope. In addition, CD9, CD63, and CD81 (ABclonal Technology, Wuhan, CHN) were measured by western blot.
In vitro proliferation assay
To examine and compare the effects of GMSC-Exo and GMSC on the proliferation of CD4+ T cells in vitro, we applied CFSE Cell Division Tracker Kit (Biolegend, San Diego, CA). Mouse CD4+ T cells were isolated from splenocytes by a CD4+ T Cell Isolation Kit (Miltenyi Biotec, Auburn, CA) according to the manufacturer’s protocol and labeled with CFSE, then cultured alone or with GMSC-Exo at the ratio of 10 ug/40 ug/90 ug: 106 (GMSC-Exo: CD4+ T cells), or with GMSC at the ratio of 1:1, 1:10 or 1:50 (GMSC: CD4+ T cells). The anti-CD3 (5 ug/ml, Biolegend, San Diego, CA) was pre-coated on the bottom of culture plate and the anti-CD28 (5 ug/ml, Biolegend, San Diego, CA) was added with GMSC-Exo or GMSC. The GMSC-Exo/GMSC were added for 3 days, then the cells were analyzed using a BD FACSCelesta flow cytometer and the levels of Th1, Th17, and Treg cytokines (IFN-γ, IL-17A, and IL-10) were detected by ELISA according to the manufacturer’s protocols [31, 32].
Induction and treatment of CIA
For CIA induction, bovine type II collagen (CII; Chondrex, Redmond, WA) was emulsified with complete Freund's adjuvant (CFA) containing 4 mg/ml heat-denatured mycobacterium (Chondrex, LLC, Seattle, WA) at a ratio of 1:1 (100 ug/mouse) and then injected intradermally into the tail (1.5 cm from the base) of DBA1/J mice to induce CIA, followed by a booster immunization with 100 ug CII in incomplete adjuvant (IFA, Chondrex, Redmond, WA) . At the moment of the boost (day 21), GMSC (1×106/mouse) and GMSC-Exo (165 ug/mouse) were injected into CIA mice (n = 6) via the lateral tail vein. In the control group, mice received equal PBS injection (n = 6).
Mice were monitored twice weekly for signs of arthritis based on arthritis scores and paw swelling. Each paw was evaluated and scored individually using a 0 to 4 scoring system: 0 = no damage; 1 = paw with detectable swelling in a single digit; 2 = paw with swelling in more than one digit; 3 = paw with swelling of all digits and instep; and 4 = severe swelling of the paw and ankle . At the day 56, all mice were euthanized and peripheral blood, spleen, lymph nodes (LN), and limbs were collected for further studies.
Histology and immunohistochemical staining
The excised paw was fixed in 4% paraformaldehyde, decalcificated, gradually dehydrated, embedded in paraffin, and sliced into 4-µm-thick sections. The sections were treated with hematoxylin and eosin (H&E) staining and immunohistochemical staining. For immunohistochemical staining, the tissue sections were depleted of endogenous peroxidase activity with methanolic H2O2, blocked with normal serum for 30 min, and incubated overnight at 4℃ with primary antibody (NF-κB P65, Cell signaling, #8242T and GAPDH, Absin Bioscience, abs100005), then incubated with HRP labeled secondary antibody (G1213, goat anti-rabbit, Servicebio, Wuhan, CHN) for 50 min at room temperature, and then reacted by 3,3′-diaminobenzidine (G1211, immunohistochemical kit DAB chromogenic agent, Servicebio, Wuhan, CHN). A negative isotype-matched control was included in each run. The sections were counterstained with haematoxylin, and the slides were scaned by PerkinElmer Pannoramic MIDI Digital Slide Scanner (3D HISTECH Ltd, Hungary). The nucleus of hematoxylin stained is blue, and the positive expression of DAB is brownish yellow. The area of brownish yellow in different groups was determined by counting four random fields per section using CaseViewer 6.0 software and histochemistry scores were calculated [35, 36].
The excised paw was fixed in 4% paraformaldehyde. The high-resolution micro-computed tomography (micro-CT) system (Skyscan 1276, Bruker Micro-CT, Kontich, Belgium) was used to acquire imaging of the three-dimensional bone. The scans were performed with entire single mouse foot having the following parameters: 20 µm voxel size at 55 kV, 200 µA, 386 ms integration time, 516 image slices. Image reconstruction was performed using graphics processing unit (GPU)-based reconstruction software, GPU-Nrecon. Ring artifact and beam-hardening correction were performed with this software as well. Reconstructed cross-sections were reoriented and ROI was further selected. Volumes of metatarsophalangeal joint were used to quantify bone erosion as follows. Second through fourth metatarsal and phalangeal bones were segmented from others using a consistent image intensity threshold. Three volumes of interest were set with ± 1 mm length in the distal and proximal direction from the center of each metatarsophalangeal joint. These volumes of interest were oriented consistently based on the 3D longitudinal axis of the third metatarsal. The bone volumes of the three metatarsophalangeal joints were then calculated.
Flow cytometry analysis
For GMSC phenotype identification, antibodies directed against human CD34, CD44, CD45, CD73, CD90, CD105, and isotype IgGs were obtained from Biolegend (San Diego, CA). The single cell suspensions were stained according to the protocol provided by the manufacturer. For cell analysis, total splenocytes or lymphocytes collected from spleen or LN of CIA mice were incubated in PBS containing 0.2% bovine serum albumin (BSA) and 1 µL of fluorophore-conjugated antibodies (0.5 mg/mL) specific for CD4, CD25, or respective isotype controls (Biolegend, San Diego, CA) at 4 ℃ for 20 min. For intracellular staining, cells were stimulated with PMA (50 ng/mL), ionomycin (1µg/mL), and brefeldin A (10 µg/mL) (Biolegend, San Diego, CA) for 4 h. Then, the cells were fixed and permeabilized with Perm/Fix solution (BD Bioscience, San Jose, CA). Finally, 1 µL of anti-IFN-γ/anti-IL17A for Th1 and Th17 subsets, and Foxp3 for regulatory T cells (Tregs) was added for 30 min in the dark. After washing with buffer, cells were analyzed using a BD FACSCelesta flow cytometer and analyzed with FlowJo software (version 10.0).
The peripheral blood from CIA mice was stand for 20 min and centrifuged at 1,000 g for 10 min to obtain the serum. The levels of cytokines (IFN-γ, IL-17A, IL-10, IL-6, and TNF-α) in culture supernatants or serum were quantified by enzyme-linked immunosorbent assay (ELISA) kit (Biolegend, San Diego, CA) according to the manufacturer’s instructions.
Western blot analysis
Total proteins were extracted from the paw of CIA mice and their concentration was determined using a BCA assay (Beyotime, Nanjing, CHN). Protein samples were applied and separated on 10% NuPAGE gel (Invitrogen, New York, NY), followed by transferring onto PVDF membranes (Millipore Inc., Darmstadt, Germany). Membranes were blocked in 5% non-fat dry milk and 0.1% Tween-20 for 1 h, followed by incubation overnight with primary antibody against mouse NF-κB p50 (Abcam, Cambridge, UK, ab32360), NF-κB p65 (Cell signaling, Danvers, MA, #8242T), Act 1 (Santa Cruz Biotechnology, Santa Cruz, CA, sc-100647), TRAF6 (Immunoway, Suzhou, CHN, YT4720), IL-17RA (Absin Bioscience, Shanghai, CHN, abs140681), and GAPDH (Absin Bioscience, Shanghai, CHN, abs100005) diluted at 1:1000 in blocking solution. Next, the HRP-conjugated secondary antibody (Absin Bioscience, Shanghai, CHN) at a dilution of 1:8,000 was used to incubate the membranes for 2 h. Immunoreactive proteins were visualized using Tanon high-sig ECL Western blotting Substrate and Tanon 5200 multifunction laser-scanning system (Tanon, CHN).
Equal amounts of proteins from the paw of CIA mice were separated on SDS-PAGE. The bands from SDS-PAGE with temperature dependence were excised and incubated with a solution of 1% SDS and 1% b-mercaptoethanol at room temperature for 1 h. The excised gel stripes were dehydrated, reduced, alkylated, and digested with trypsin at 37 ℃ for 16 h. Then, the peptides were delivered onto a nano RP column and eluted with gradient (50–80%) acetonitrile (ACN) for 60 min at 400 nL/min. After that, the elute was injected into a Q-Exactive mass spectrometry set in a positive ion mode and a data dependent manner with a full MS scan from 350 to 2,000 m/z. High collision energy dissociation was employed as the MS/MS acquisition method [37, 38]. Proteome Discoverer 1.4 (Thermo Fisher Scientific, Waltham, MA, USA) was used to convert raw MS/MS data into a MGF format. The exported MGF files were searched with Mascot v2.3.01 against the uniprot 20180305 mouse database (17,021 sequences) with a typtic specificity, allowing two missed cleavage. Carbamidomethylation was considered as fixed modification whereas oxidation (M) and Gln->pyro-Glu (N-term Q) as variable modifications. The mass tolerances for MS and MS/MS were 15 ppm and 20 mmu, respectively. Proteins with false discovery rates < 0.01 were further analyzed.
SPSS 16.0 was used to perform statistical analysis. Significance was assessed by one-way analysis of variance (ANOVA) followed by LSD post hoc test. P values less than 0.05 were considered as significant.