MSCs relieve SLE by modulation of Th17 cells through MMPs–CCL2–CCR2–IL-17 pathway

2.1 Patients and mice

All patients fit for the classification criteria of SLE established by American College of Rheumatology (ACR) in 1982. Their clinical features were shown in Table S1. After approved by medical ethics committee of Drum Tower Hospital (No. 2008017) and got informed consent from SLE patients and healthy control, the bone marrow and anticoagulant venous blood were collected.

Sixteen-week-old female C57BL/6 mice and female MRL/lpr mice (lupus mouse model) were purchased from Beijing Experimental Animal Center of Military Medical Sciences. All mice were raised in SPF grade animal housing. MRL/lpr mice were randomly divided into three groups: umbilical cord-derived MSCs (UC-MSCs) transplantation combined with IgG2b isotype control group (n = 6), UC-MSCs transplantation combined with CCL2 antibody group (n = 6) and untreated group (n = 6). C57BL/6 mice were used as negative control. 1 × 10⁶ UC-MSCs were injected into 16-week-old MRL/lpr mice through tail vein. In the same time, 25 µg CCL2 neutralizing antibody or IgG2b was injected intraperitoneally every week. All mice were sacrificed at 29-week-old.

2.2 Reagents

Culture medium including DMEM-F12, RPMI 1640, fetal bovine serum (FBS) were purchased from Gibco. Lymphocyte separation liquid was bought from TBD Company. To purify mouse naïve T lymphocytes, immune magnetic bead was gained from Miltenyi. Antimouse-CD3, antimouse-CD28, recombinant mouse IL-1β, IL-6, IL-21, IL-23, and TGF-β1 were from Santa Cruz Biotechnology. Recombinant human MMPs and CCL2 were purchased from Peprotech. Fluorescent antibody of CD4-FITC, IL-17-APC, CCR2-PE, and isotype control were from eBioscience. Enzyme-linked immunosorbent assay (ELISA) kits for IL-17, MMP1, MMP3 and CCL2 were from R&D System. Luminex kits of MMPs were bought from Merch & Millipore Company.

2.3 Isolation and culture of UC-MSCs

Fresh umbilical cord originated from informed healthy mothers in maternity department after normal deliveries was placed in balanced salt solution. After blood eliminating and vessel were removed, Wharton's jelly was separated, washed and cut into 1–4 mm³ tissue. Cells (1 × 10⁶/cm²) were incubated in in low sugar DMEM medium with 5% FBS at 37°C and 5% CO₂. Culture medium was removed 1 week later and replaced by complete medium every 3 days. When adherent cells reached 60%–80% fusion, they were digested with 0.25 g/L trypsin and passaged (1 × 10⁴/cm²). Flow cytometry (FCM) was used to identify the surface markers of stem cell including CD106, CD105, CD44, and CD29. Their positive percentage was ≥95%. And the percent of CD45, CD34, CD14, CD79 and HLA-DR is ≤2%. Second to fifth generation of UC-MSCs were used for transplantation.

2.4 Isolation and culture of human bone marrow-derived MSCs (BM-MSCs)

Ten milliliter heparin-anticoagulated bone morrow from SLE patients and healthy controls was diluted with 0.9% saline and added onto same volume of lymphocyte separation liquid, then centrifuged 400g for 20 min at 4°C. Mononuclear cells were collected, washed and cultured in bottle (5 × 10⁶cells/ml) in complete DMEM medium with 10% FBS at 37°C and 5% CO₂. One week later, the culture medium was changed and non-adherent cells were removed. After that, culture medium was replaced by low sugar complete DMEM every 3 days. When adherent cells were more than 90%, they were harvested by digestion using 0.25% trypsin and passaged according to the proportion 1:2. Amplified BM-MSCs were counted and identified.

2.5 Isolation of naïve T lymphocytes from mouse spleen

Naïve T lymphocytes were isolated by magnetic-activated cell sorting (MACS). Mononuclear cells of mouse spleen were suspended in 40 μl magnetic buffer and 10 μl CD4⁺ T cell biotin-antibody cocktail II MicroBeads, then incubated at 4°C for 15 min. After incubation with anti-Biotin MicroBeads for 15 min at 4°C, the cells were washed, centrifuged 400g for 10 min and resuspended. Cells suspending liquid was added onto LS column. Cells were collected from LS column after sorting and centrifuged 400g for 10 min, incubated with CD62L microbead at 4°C for 15 min. Then the cells were washed and added into MS column. Positive cells were collected.

2.6 Coculture of UC-MSCs with Th

The three to five generations of UC-MSCs were incubated overnight in lower chamber of 24-hole-transwell (10⁴cells/well) in DMEM-F12 medium. Naïve T cells were cultivated in RPMI 1640 with Th17 stimulation agents: anti-mCD3 (2 μg/ml), anti-mCD28 (2 μg/ml), IL-1β (10 ng/ml), IL-6 (30 ng/ml), IL-21 (50 ng/ml), IL-23 (20 ng/ml), and TGF-β1 (3 ng/ml) in the upper chamber (10⁴ to 10⁶cells/well) for 72 h. According to different groups, various reagents were added: anti-human-CCL2-Ab, IgG2b isotype control, recombinant human CCL2, MMP-processed CCL2 (mpCCL2), and recombinant human MMPs, respectively.

2.7 Detection of IL-17 and CCR2 expression on T lymphocytes

The expression of CD4, IL-17 and CCR2 on T lymphocyte were measured by flow cytometry. Antibody staining was performed according to manufacturer's instruction. Peripheral blood mononuclear cells (PBMC) separated by Ficoll were washed and added into 80 µl RPMI 1640 culture medium with 10 µl PMA, 10 µl Ionomyocin, 5 µl BFA in 37°C, 5% CO₂ for 4 h. After washing, the cells were stained by FITC-CD4 and CCR2-PE darkly at room temperature (RT) for 15 min, then were membrane-broken by agent A/B for 15 min, stained by IL-17-APC for 30 min. The washed stained cells were resuspended and tested.

2.8 Detection of the concentration of cytokines

The concentration of CCL2, MMPs, IL-17, IgG, anti-nuclear antibody (ANA) and anti-double-stranded DNA (dsDNA) in serum and supernatant of BM-MSCs were detected by ELISA and Luminex according to manufacturer's instructions. The procedure of Luminex: assay buffer was added to pre-wet and closing, then microtiter plate was placed on a shaker for 10 min. Assay buffer was removed. Standard set and suitable substrate solution was added. 25 µl diluted sample and 25 µl pre-mix microbead were added into every well. The plate was covered with special plastic film board and shook at 4°C for 18–20 h. After washed twice, the sample was incubated with secondary antibody (25 µl/well) at RT for 2 h. Then 25 µl Streptavidin-Phycoerythrin was joined and incubated at RT for 30 min. The content of inner panel was removed. Sheath fluid was added to each well and shook for 5 min. The board was placed in Luminex 100™ IS, 200™ to analyze.

2.9 Generation of MMP-processed CCL2

To generate MMP-processed CCL2 (mpCCL2), we added 10 ng pure recombinant human MMP (rhMMP) directly to 50 μg pure recombinant human CCL2 (rhCCL2) at 37°C for 4 h. Mp-CCL2 was directly used in assay.

2.10 Detection of the kidney of lupus mice

Twenty-four-hour urine was collected using metabolism cages. Urine protein was measured by Coomassie Blue staining method. The kidneys of MRL/lpr mice and C57BL/6 mice were cut into small pieces and fixed in 10% formalin at 4°C for 24 h. Paraffin sections (4 µm) of renal tissues were stained with hematoxylineosin (HE). The presence of immune complex in kidney was detected by immune fluorescent assay.

2.11 Statistics analysis

Data are expressed as mean ± standard deviation. Data analysis was done using two-tailed Mann–Whitney and Wilcoxon t test by SPSS 13.0 software. p Value < 0.05 was considered as statistically significant.

3.1 The level of CCL2-CCR2-Th17-IL-17 increased in SLE

The percentage of IL-17⁺CD4⁺ T lymphocytes (Th17) (3.077 ± 0.515% vs. 0.223 ± 0.084%, p < 0.0001) and the concentration of IL-17 in peripheral blood of SLE patients (7.56 ± 1.69 pg/ml vs. 3.62 ± 0.42 pg/ml, p < 0.05) was significantly higher than health control (shown in Figure S1). The concentration of chemokine-CCL2 in serum and supernatant of BM-MSCs (Figure 1) and the expression of CCR2 on Th17 cells in blood (Figure 2) were obviously increased in SLE patients. It is also found that the concentration of IL-17 (73.13 ± 24.67 pg/ml vs. l3.14 ± 1.14 pg/ml, p < 0.001) and CCL2 (498.0 ± 175.6 pg/ml vs. 69.29 ± 15.53 pg/ml, p < 0.01) in the serum, and the percentage of Th17 cells (1.06 ± 0.17% vs. 0.42 ± 0.14%, p < 0.05) and CCR2⁺ Th17 cells (26.59 ± 8.62% vs. 18.94 ± 5.12%, p > 0.05) in PBMCs of MRL/lpr mice were higher than those of C57BL/6 mice. These results displayed that CCL2-CCR2-Th17-IL-17 was involved in SLE.

3.2 MSCs inhibit IL-17 production through CCL2–CCR2 pathway to relieve SLE

We further investigated the effects of MSCs and CCL2 on SLE disease. Compared with normal control mice, the proteinuria, serum ANA concentration and immune complex in kidney were all significantly higher in MRL/lpr mice with disease activity of lupus (Figure 3). MSCs transplantation (MSCs + IgG2b isotype control group) evidently decreased these indexes especially for urine protein (p < 0.05), ANA concentration (p < 0.01), and immune complex in kidney of MRL/lpr mice (Figure 3). It is interesting that antimouse-CCL2-Ab strengthened effects of MSCs on lupus mice (Figure 3), suggesting that CCL2 participate in the immune-regulation function of MSCs in SLE.

Next, to explore if MSCs alleviate SLE through CCL2–CCR2–Th17 pathway, the levels of IL-17, CCL2 and CCR2 after UC-MSC transplantation in lupus mice were detected. The serum IL-17 and CCL2, Th17 cell percentage and CCR2 expression on PBMCs in lupus mice were all decreased by UC-MSCs (Figure 4). On the other hand, the effects of UC-MSCs combined with CCL2-Ab were much stronger than UC-MSCs along, though without statistic difference (Figure 4). These results confirmed the modulation of MSCs on CCL2-CCR2–Th17–IL-17 pathway.

Then coculture experiment was carried out to further identify the molecule pathway of MSCs on Th17 cells. UC-MSCs were cocultured with spleen CD4⁺ Th cells from lupus mice in transwell system for 72 h with different ratio (UC-MSC: Th cells = 1:10, 1:50, 1:100). Then IL-17A expression on CD4⁺ Th cells was detected. It is found that UC-MSCs could significantly inhibit the production of IL-17A especially in the group of ratio 1:10 (Figure 5). However, IL-17A expression was increased by anti-human-CCL2-Ab in coculture system of MSCs and Th17 cells (Figure 5). CCL2-Ab reversed the inhibition of UC-MSCs on Th17 cells, suggesting that CCL2 was crucially involved in the effect of MSCs on Th17 cells.

3.3 MSCs secrete MMPs to reverse CCL2 function and modulate Th17 cells

We next explore how CCL2 modulates the control of MSCs on Th17 cells. MSCs could produce MMPs. It is found that the level of MMP2/3/8 in the serum and supernatant of BM-MSCs of SLE patients were evidently elevated than normal controls (Figure S2). Similarly, in MRL/lpr mice, the concentration of MMP2/8/9 in the serum were also much higher than those of C57BL/6 mice, and transplantation of UC-MSCs obviously decreased them (Figure 6). Therefore, MMPs may participate in SLE and normal MSCs could relieve SLE through secreting MMPs.

To explore the modulation of MMPs on CCL2–CCR2–Th17–IL-17 axis, recombinant human CCL2 (rhCCL2, 100 ng/ml), rhMMPs (10 ng/ml), MMP-hydrolyzed CCL2 (mpCCL2, 100 ng/ml) and UC-MSCs were added into the culture system of naïve T cells of lupus mouse respectively. RhCCL2 and rhMMP improved, but mpCCL2 and UC-MSCs decreased the percentage of CCR2⁺ IL-17⁺ T lymphocytes (see Figure S3), suggesting that the hydrolysis of CCL2 by MMP reverse its function on Th17 cells.