Human Umbilical Cord Mesenchymal Stem Cell-derived Exosomes Induce Macrophage M2 Polarization by Antagonizing LPS-mediated Stimulation of the NF-κB and STAT3 Pathways

Previous methods were followed for the isolation and propagation of hucMSCs [18]. These cells (third passage) were differentiated into osteoblasts and adipocytes using respective induction media (Gibco Grand Island, NY, USA). Alizarin Red and Oil Red O stains were used to observe osteogenic and adipogenic differentiation. Exosomes (hucMSCs-EX) derived from a conditioned medium of hucMSCs (hucMSCs-CM) were isolated and purified as per the established protocols [19]. Briefly, hucMSCs were grown to 60% confluence, and were rinsed three times with PBS to remove any particle of fetal bovine serum (FBS, Invitrogen). Cells were continuously grown without serum and exosomes in RPMI-1640 medium (Invitrogen) for 48 h. This exhausted medium is considered a conditioned medium derived from hucMSCs (hucMSCs-CM). The medium was collected and distributed into two parts: one part of hucMSCs-CM was filtered via a 0.22 μm syringe filter (Millipore, USA) and kept at -80°C for further use. The second part of hucMSCs-CM was centrifuged (1000 × g, 20 min) to discard cell debris, followed by two re-centrifugations (2,000 × g and 10,000 × g, 20 min each). The supernatants were purified (100 kDa MWCO [Millipore, USA], 1,000 × g, 30 min). The purified mixture was placed on a cushion of 5 mL of 30% sucrose/D2O and ultracentrifuged using a Beckman Coulter Optima L-90K ultracentrifuge (100,000 × g, 90 min, 4°C) [11]. The fractions enriched with microvesicles were obtained and then mixed with PBS. This was followed by three centrifugations (1,000 × g, 30 min) with 100 kDa MWCO. Precipitates were resuspended in sterile PBS and centrifuged (3000 rpm, 10 min, 4°C). After adsorption of 20 μL of the purified hucMSCs-EX onto copper grids and incubation at 25°C for two min, the grids were stained for 10 min with 30 μg/L phosphotungstic acid (pH 6.8) at 25°C [11]. The sample was dried at 25°C and then examined via transmission electron microscope (TEM) (Olympus, Tokyo, Japan). The size and dispersion of hucMSCs-EX were quantified and examined using Nanosight NTA and the Zetasizer-Nano software. Finally, the purified hucMSCs-EX were extracted, filtered (0.22 μm), and then kept at -80°C for further use. Bichicondinic acid (BCA) was employed to determine the protein concentration of the isolated exosomes. The resulting value was then used to calculate the exosomes' concentration. Approximately 50 μg/mL exosomes were preserved at -80°C until use. Fresh umbilical cord tissues were collected from full-term newborn delivered through cesarean section after obtaining written consent from their parents [18]. The Ethics Committee of Bengbu Medical University, Bengbu, China, approved all the relevant experimental protocols.

The cell line (murine macrophage RAW264.7) was received from the Cell Bank of Shanghai, China. Cells in logarithmic growth were grown in RPMI-1640 with 10% FBS in the presence or absence of LPS (1 μg/mL) (5164948, Hangzhou, China) for 4 h. After incubation, cells were rinsed with PBS (thrice) and maintained in a complete medium (control), hucMSCs-CM (1:1 diluted while using a fresh medium), and 5 μg/mL hucMSCs-EX for 48 h in a 6-well dish. The propagated macrophages and resultant supernatants were obtained and stored for further experiments.

Total RNA contents were isolated from respective cells via TRIzol reagent (Invitrogen). The cDNAs were reverse-transcribed as per the manufacturer's directions via a reverse transcription kit (Invitrogen). The qPCR reactions were conducted using primers that were specific to mice and targeted the TNF-α, IL-10, IL-6, IL-1β, Arg-1, and β-actin genes (Table 1). mRNA expression was normalized with β-actin, which is regarded as the internal or standard control. The Bio-Rad Real-Time System was used to conduct PCR reactions via TB GreenTM Premix Ex TaqTM II (Tli RNaseH Plus).

As per the manual's instructions (Dakewe, Beijing, China), the quantity of IL-6 and IL-10 in the collected supernatants was monitored using ELISA kits. The absorbance (OD) was detected at 450 nm via the NanoDrop™ 8000 Spectrophotometer (ND-8000-GL, Thermo Scientific™).

The lysates of total or nuclear proteins were isolated from cultured cells and hucMSCs-EX. Proteins contents were quantified with a BCA kit (Thermo Fisher Scientific). Protein in aliquots of equal weight (40 μg) was loaded onto a 10% SDS-PAGE and subsequently transferred to nitrocellulose membranes. After 1 h of blocking in 5% skim milk at 25°C, the blots were kept at 4°C for 24 h with respective primary antibody dilutions and then labeled with secondary HRP-linked antibody at 37°C for 2 h. The bands of protein were detected via an HRP substrate (EMD Millipore). and examined using MD ImageQuantTM Software (G: Box; Syngene, Cambridge, UK). β-actin, as well as histone H3.1, were employed as reference controls for cell cytoplasmic and nuclear proteins. Following were the Rabbit monoclonal antibodies used for WB analysis: anti-human CD9 antibody (1:500: BS60359, Bioworld Technology, USA), anti-human CD63 (1:1000: ab134045, Abcam, Cambridge, UK), anti-human Alix 3 (1:1000: ab186728, Abcam), anti-Lamin A antibody (1:2000: ab108595, Abcam), anti-human β-Tubulin (1:2000: AP0064, Bioworld), anti-mouse CD206 (1:1000: ab64693, Abcam), anti-mouse NF-κB p65 (1:1000: #C48676↗, SAB, CA), anti-mouse NF-κB p-p65 (S536, 1:1000: ab76302, Abcam), anti-mouse p-STAT3 (Tyr539, 1:500, #13150, SAB), STAT3 (1:500, #53229, SAB), Rabbit polyclonal anti-mouse Histone H3.1 (1:500: BS90644(P68433↗), Bioworld) and anti-β-actin (1:2000: AP0060, Bioworld). The goat anti-Rabbit IgG(H+L)-HRP antibody (1:2000, Bioworld Technology Cat# BS13278, RRID: AB_2773728) was purchased from Bioworld.

Cells were cultured for 24 h on glass slides coated in 24-well plates. Further, they were co-incubated with complete RPMI-1640 medium, hucMSCs-CM, and hucMSCs-EX for 48 h. Cells were rinsed twice with PBS and fixed in 4% paraformaldehyde (PFA, AR-0211; DingGuo Biotechnology China) solution for 30 min at 4°C. Blocking (5% BSA) was performed and cells were kept with rabbit anti-mouse CD206 monoclonal antibody (1:500, ARG22456↗, Arigobio) at 4°C for 24 h and followed by FITC-linked anti-rabbit secondary antibodies (1:200, ARG23840↗, Arigobio) at 37°C for 2 h. Cells were stained with DAPI to detect cell nuclei. Fluorescence images of cells were obtained using ECLIPSE Ti-S, Nikon, Japan.

Metabolic or proliferative activity of RAW264.7 cells was detected via the MTT assay. Precisely, approximately 2 × 103 cells/well were grown in a 96-well plate and kept with 100 μL of complete medium (serve as the control group), hucMSCs-CM and hucMSCs-EX mixed in the presence or absence of LPS (1 μg/mL). The cells were subjected to incubation at 25°C for 4 h after adding 20 μL MTT to each well at the respective times of 24, 48, and 72 h. Finally, the media were discarded and absorbances were monitored at 450 nm via a microplate reader (RNE90002↗, USA).

In this assay, RAW264.7 cells (5 × 104 in 200 μL of incomplete medium) were allowed to grow in a 24-well plate (8.0-µm polycarbonate membrane) (Costar, CA, USA). They were seeded to the upper surface of the compartment while the lower surface was filled with only medium (600 µL). Further, hucMSCs-CM and 600 µL hucMSCs-EX were mixed, respectively. In each well, LPS (1 μg/mL) was added. The upper surface of the membrane was cleared of any residual cells after a 12-h incubation via a cotton swab. The migrated cells were fixed for 8 min with 4% PFA on the lower surface of the membrane. Crystal violet solution was used to stain the cells for 15 min. After washing with PBS (thrice) cells were dried, examined, and quantified under an inverted microscope (Olympus, Tokyo).

Data was statistically evaluated via GraphPad 9.3 (Graph Pad Software) and SPSS 21.0 (IBM Corp.) software and were depicted as mean ± SEM derived from ≤ triplicate analyses. Group comparison was performed using Student’s t-test and One-way ANOVA with Tukey’s post hoc tests. A p-value ≤ 0.05 was regarded as a significant threshold.

This study was performed on cell lines and on human umbilical cord tissue. Human umbilical cords used in this study are normally discarded as medical waste so no ethical concerns are involved in obtaining hucMSCs-EX. There were no humans were directly involved in this study. No work with human subjects was directly involved in our study.

Primary cultured hucMSCs adhered to the plastic surface after migrating from the human umbilical cord tissues. They have fibrous morphology and fingerprint-like growth characteristics under an inverted microscope (Fig. 1A1 and A2). In the third passage, after subcultures, a population of hucMSCs that was relatively homogeneous was observed (Fig. 1A3). Multidirectional in vitro differentiation (adipogenic and osteogenic) was displayed by hucMSCs. As shown in Fig 1B2, the cells comprised fat particles and were stained red; similarly, calcium deposits were stained red in the cells (Fig. 1B3). Exosomes were predominantly round lipid-coated microvesicles resembling tea saucers, as detected using TEM. The exosomes showed a range of sizes from 40 to 100 nm (Fig. 1C and D). The Western blot (WB) analysis revealed an upregulation of exosome proteins CD9, CD63, and Alix3 in the hucMSCs-EX (Fig. 1E). The extraction of hucMSCs-EX was successful.

Compared with RPMI-1640 medium, hucMSCs-CM or hucMSCs-EX significantly increased Arg-1 and IL-10, while markedly reducing TNF-α and IL-6 expression in LPS-induced cells (Fig. 2A). Further, IL-6 and IL-10 were identified in the culture supernatants of the macrophages via ELISA, as previously outlined. In line with the qRT-PCR findings, hucMSCs-EX or hucMSCs-CM substantially inhibited the secretion of IL-6 (Fig. 2Ba), an essential pro-inflammatory cytokine, while markedly increasing the release of IL-10 (Fig. 2Bb).

The protein expression of CD206 (mannose receptor), a widely known marker for M2 macrophages was detected to identify whether hucMSCs-EX elicited the M2 phenotype [20], in cells via FC and WB analysis. The hucMSCs-EX or hucMSCs-CM substantially raised the quantity of CD206 expressing macrophages (Fig. 3A). In consensus with the findings from FC, immunofluorescence staining and WB analyses revealed substantially increased levels of CD206 in respective cells exposed with hucMSCs-CM and hucMSCs-EX, relative to the control group (Fig. 3B and C).

To identify whether hucMSCs-EX or hucMSCs-CM can promote macrophage proliferation and migration, MTT and cell counting assays were carried out. As depicted in Fig. (4A), the OD values of hucMSCs-EX and hucMSCs-CM groups were higher than those in the control and LPS-induced RAW264.7 groups 48 h after treatment, respectively. In line with the findings of the MTT, the cell counting assay demonstrated that hucMSCs-EX or hucMSCs-CM markedly enhanced proliferation in LPS-induced cells 48 h after treatment (Fig. 4B). The migratory capability of respective cells after treatment with hucMSCs-EX was then evaluated. The data depicted in Fig (4C and D) suggest that 12 h after treatment, the quantity of migrated cells was higher in the hucMSCs-EX or hucMSCs-CM group in contrast to the control group.

Western blotting determined the concentrations of the cytoplasmic protein NF-κB p-p65 and the nuclear protein NF-κB P65 in respective cells. As illustrated in Fig. (5A and B), the distribution of NF-κB p65 proteins in uninduced RAW 264.7 cells was predominantly found in the cytoplasmic fraction. An increase in phosphorylated NF-κB p65 (NF-κB p-p65) and NF-κB p65 was seen after 1 h of LPS stimulation (Fig. 5A). However, this effect was antagonized by hucMSCs-EX (Fig. 5A and C). The qRT-PCR findings indicated that the NF-κB p65 activation induced by LPS specifically targeted IL-1β and TNF-α (Fig. 5D and E). The expressions of TNF-α (Fig. 5D) and IL-1β (Fig. 5E) were inhibited by hucMSCs-EX. The results demonstrated that hucMSCs-EX inhibited the LPS-promoted phosphorylation of NF-κB p65, NF-κB p65, and the downstream secretions of TNF-α and IL-1β. Considering the significance of the NF-κB and STAT3 pathways in M2 macrophage polarization [21], a possible connection between the role of MSCs-EX and STAT3 signaling was monitored. RAW264.7 cells were exposed to hucMSCs-EX at various time intervals. The WB analysis revealed a significant upregulation of p-STAT3 expression 1 h after treatment, which persisted in an activated state for the next 4 h (Fig. 6A). To confirm the function of STAT3 stimulation in the polarization induced by hucMSCs-EX, cells were treated with hucMSCs-EX with or without the presence of STAT3 inhibitor, S3I-201, under both normal and LPS-induced environments. The findings indicated that the upregulation of Arg-1 and the activation of p-STAT3 by hucMSCs-EX were both inhibited by S3I-201 (Fig. 6B and C). Further, S3I-201 inhibited the synthesis of CD206 protein in cells induced by hucMSCs-EX (Fig. 6D). Collectively, these results demonstrate that MSCs-EX stimulates the STAT3 pathway in respective cells to induce M2 polarization.

We declare that all data and supporting information is provided in this article.