Phorbol ester activates human mesenchymal stem cells to inhibit B cells and ameliorate lupus symptoms in MRL. Faslpr mice

Abstract

Rationale: Systemic lupus erythematosus (SLE) is a multi-organ autoimmune disease characterized by autoantibody production by hyper-activated B cells. Although mesenchymal stem cells (MSCs) ameliorate lupus symptoms by inhibiting T cells, whether they inhibit B cells has been controversial. Here we address this issue and reveal how to prime MSCs to inhibit B cells and improve the efficacy of MSCs in SLE. Methods: We examined the effect of MSCs on purified B cells in vitro and the therapeutic efficacy of MSCs in lupus-prone MRL.Fas^lpr mice. We screened chemicals for their ability to activate MSCs to inhibit B cells. Results: Mouse bone marrow-derived MSCs inhibited mouse B cells in a CXCL12-dependent manner, whereas human bone marrow-derived MSCs (hMSCs) did not inhibit human B (hB) cells. We used a chemical approach to overcome this hurdle and found that phorbol myristate acetate (PMA), phorbol 12,13-dibutyrate, and ingenol-3-angelate rendered hMSCs capable of inhibiting IgM production by hB cells. As to the mechanism, PMA-primed hMSCs attracted hB cells in a CXCL10-dependent manner and induced hB cell apoptosis in a PD-L1-dependent manner. Finally, we showed that PMA-primed hMSCs were better than naïve hMSCs at ameliorating SLE progression in MRL.Fas^lpr mice. Conclusion: Taken together, our data demonstrate that phorbol esters might be good tool compounds to activate MSCs to inhibit B cells and suggest that our chemical approach might allow for improvements in the therapeutic efficacy of hMSCs in SLE.

Keywords: B cell; CXCL10; PD-L1; mesenchymal stem cell; phorbol ester; systemic lupus erythematosus.

Figure 1

Effects of mMSCs on the proliferation of and IgM production by mB cells. (A-B) mMSCs (0.1-3 × 10⁴ cells/well) and MRL.Fas^lpr mB cells (1 × 10⁵ cells/well) were co-cultured for 72 h. LPS (1 μg/mL) was used to activate mB cells. The proliferation of and IgM production by mB cells were measured by the mitogen assay (A) and ELISA (B), respectively. (C-D) mMSCs (1 × 10⁴ cells/well) were added to the upper (U) or lower (L) wells of transwell plates and mB cells (1 × 10⁵ cells/well) to the L wells. After incubation with LPS for 72 h, the mitogen assay (C) and ELISA (D) were performed. (E) The levels of TGF-β, IL-10, and PGE₂ accumulated in the culture medium of BM cells and mMSCs for 24 h were measured by ELISA. NO level was measured with Griess reagent. (F) Expression levels of COX-2, iNOS, and TGF-β mRNAs in BM cells and mMSCs were assessed by RT-PCR. ^*p < 0.01 (n = 3).

Figure 2

Effects of mMSCs on the migration of mB cells. (A) Expression levels of chemokine mRNAs in BM cells and mMSCs were assessed by RT-PCR. (B) The levels of CCL2, CCL4, CCL5, and CXCL12 accumulated in culture medium of BM cells and mMSCs for 24 h were measured by ELISA. (C) Expression levels of chemokine receptors and their mRNAs in mB cells were assessed by western blotting and RT-PCR, respectively. (D) mMSCs were transfected with negative (neg) control, CCL2, or CXCL12 siRNA for 48 h. The levels of CCL2 and CXCL12 accumulated in culture medium for 24 h were measured by ELISA. (E-F) CMFDA-labeled mB cells (1 × 10⁵ cells/well) were added to the upper wells of transwell plates with a 5-μm insert. mMSCs (0.3-3 × 10⁴ cells/well), which had been transfected with negative control, CCL2-siRNA, or CXCL12-siRNA, were added to the lower wells (E). CMFDA-labeled mB cells were pre-treated with dimethyl sulfoxide (0.1%, Control), CCR2 antagonist RS102895 (30 μg/mL), or CXCR4 antagonist AMD3100 (300 μg/mL) for 1 h, washed three times, and added to the upper wells. mMSCs (0.3-3 × 10⁴ cells/well) were added to the lower wells (F). After 1.5 h, the number of CMFDA-labeled mB cells migrating to the lower well was determined. (G) For time-lapse imaging, mMSCs (70 μL of 0.3 × 10⁶ cells/mL) were seeded into the left chamber and mB cells (70 μL of 1 × 10⁶ cells/mL) into the right chamber of culture-insert μ-Dish^35mm culture dishes. Images were acquired every 2 min for 12 h after 1-h pre-incubation. Representative photos are shown (n = 3). The numbers of mB cells passing through the white boxes are shown. ^*p < 0.01 (n = 3).

Figure 3

Effects of PMA-hMSCs on IgM production by hB cells. (A) hMSCs (0.1-3 × 10³ cells/well) and hB cells (1 × 10⁵ cells/well) were co-cultured for 72 h. (B) hMSCs were treated with PMA (10 ng/mL) for 24 h and washed three times with medium. PMA-treated hMSCs (PMA-hMSCs; 0.1-3 × 10³ cells/well) and hB cells (1 × 10⁵ cells/well) were co-cultured for 72 h. (C) hMSCs were treated with ingenol-3-acetate (I3A, 10 µg/mL) or phorbol 12,13-dibutyrate (PdBU, 10 µg/mL) for 24 h and washed three times with medium. Chemical-treated hMSCs (1 × 10³ cells/well) were co-cultured with hB cells (1 × 10⁵ cells/well) for 72 h. (D) hMSCs were treated with PMA in the presence or absence of the PKC inhibitor Go6983 (1 μg/mL) for 24 h. PMA-hMSCs were washed three times with medium. PMA-hMSCs (1 × 10³ cells/well) were co-cultured with hB cells (1 × 10⁵ cells/well) for 72 h. (E) PMA-hMSCs (1 × 10³ cells/well) were added to the lower wells and hB cells (1 × 10⁵ cells/well) to the upper wells of transwell plates with a 5-μm insert. CpG-oligodeoxynucleotide (ODN, 5 μg/mL) was used to activate hB cells. IgM production by hB cells was measured by ELISA (A-E). (F) Total RNA was isolated from chemically untreated hMSCs (UN) or PMA-treated hMSCs (PMA). Gene expression levels were assessed by RT-PCR. ^*p < 0.01 (n = 3).

Figure 4

Effects of PMA on the migration of hMSCs. (A) hMSCs were activated with PMA for 24 h. Chemokine expression levels were measured by RT-PCR and ELISA. UN, untreated. (B) Expression levels of chemokine receptors in hB cells were assessed by RT-PCR. (C) PMA-hMSCs were transfected with negative-control, CCL2, CXCL10, or CXCL12 siRNA for 48 h, and chemokine expression levels were analyzed by RT-PCR. PMA-hMSCs (0.03-0.3 × 10⁴ cells/well) were added to the lower wells and CMFDA-labeled hB cells (1 × 10⁵ cells/well) to the upper wells of transwell plates with a 5-μm insert. After 1.5 h, the number of CMFDA-labeled hB cells migrating to the lower well was determined. (D) For time-lapse imaging, hMSCs (70 μL of 0.3 × 10⁶ cells/mL) were seeded into the left chamber and hB cells (70 μL of 1 × 10⁶ cells/mL) into the right chamber of culture-insert μ-Dish^35mm culture dishes. Images were acquired every 2 min for 12 h. Representative photos are shown. The numbers of hB cells passing through the white boxes are shown. ^*p < 0.01 (n = 3).

Figure 5

Effects of PMA-hMSCs on the viability of hB cells. (A) hMSCs were activated with PMA for 24 h. Expression levels of the death ligands PD-L1, PD-L2, and FasL were measured by RT-PCR and flow cytometric analysis. UN, untreated. (B-C) PMA-hMSCs (1 × 10³ cells/well) were co-cultured with hB cells (1 × 10⁵ cells/well) for 72 h in the presence of blocking antibodies against PD-L1, PD-L2, or FasL (B). PMA-hMSCs, which had been transfected with PD-L1 or FasL siRNA, were co-cultured with hB cells (C). ODN (5 μg/mL) was used to activate hB cells. IgM production by hB cells was measured by ELISA assay. (D-F) PMA-hMSCs (0.1 × 10⁶ cells), which had been transfected with PD-L1 siRNA, were cultured with hB cells (1 × 10⁶ cells) in 35-mm culture dishes for 24 h. hB cells were stained with anti-CD19-APC and then stained with FITC-Annexin V and propidium iodide (PI). Cells were analyzed using a flow cytometer (D). hB cells were stained with anti-CD19-APC and Intracellular Caspase Detection ApoStat kit and analyzed using a flow cytometer (E). CellEvent Caspase-3/7 Green ReadyProbes Reagent was added to the culture of hB cells and hMSCs, and the cells were imaged every 10 min for 24 h with a Biostation IM-Q microscope (Nikon). Green fluorescent cells were considered apoptotic (F). ^*p < 0.01 (n = 3).

Figure 6

In vivo efficacy of PMA-hMSCs in MRL.Fas^lpr mice. (A-B) MRL.Fas^lpr mice were intravenously injected with PBS (control), naïve hMSCs (4 × 10⁴ cells /injection), or PMA-hMSCs (4 × 10⁴ cells/injection) once at the age of 12 weeks. Survival was measured every week (A) and body weight (B) every 2 weeks up to 30 weeks of age. ^*p < 0.01 (n = 6). (C-H) Injections were performed as in (A). Surviving mice were sacrificed at the age of 22 weeks. The serum levels of anti-dsDNA IgG (C) and total IgG (D) were measured every 3 weeks. Proteinurea levels were measured at the age of 22 weeks (E). Kidney sections were stained with antibodies against IgG and C3 complement (F). Total RNA was isolated from spleen cells and the expression levels of inflammatory cytokine genes (IL-1β, IL-12, IFN-γ, and TNF-α) were examined by RT-PCR (G). The ratios of Foxp3-expressing CD4⁺ Treg cells (CD4⁺Foxp3^int) and IgG-producing CD138⁺ plasma cells (CD138⁺IgG^int) in the spleen were measured by flow cytometry (H). ^*p < 0.01 (n = 6).

Figure 7

Immunohistochemical analysis. Kidney sections used in Figure 6F were stained with antibodies against CD3 (T cells), CD19 (B cells), F4/80 (macrophages), CD209b (dendritic cells), or Foxp3 (Treg cells).

Figure 8

Effects of PMA-hMSCs on xenogeneic mB cells. (A) hMSCs (0.03-1 × 10⁴ cells/well) and MRL.Fas^lpr mB cells (1 × 10⁵ cells/well) were co-cultured for 72 h. LPS (1 μg/mL) was used to activate mB cells. The proliferation of and IgM production by mB cells were measured by the mitogen assay and ELISA, respectively. (B) hMSCs (0.03-1 × 10⁴ cells/well) and MRL.Fas^lpr mT cells (1 × 10⁵ cells/well) were co-cultured for 72 h. Concanavalin A (ConA, 1 μg/mL) was used to activate mT cells. The proliferation of and IFN-γ production by mT cells were measured by the mitogen assay and ELISA, respectively. (C-D) hMSCs (1 × 10⁴ cells/well) were activated with PMA (3-30 ng/mL) for 24 h and then co-cultured with mB cells (1 × 10⁵ cells/well) (C) or mT cells (1 × 10⁵ cells/well) (D). (E) hMSCs were activated with PMA (10 ng/mL) for 1 to 7 days and then co-cultured with mB cells. ^*p < 0.01 (n = 3).