Effective combination of human bone marrow mesenchymal stem cells and minocycline in experimental autoimmune encephalomyelitis mice

Abstract

Introduction: Multiple sclerosis (MS) is the most common inflammatory demyelinating disorder of the central nervous system (CNS). Minocycline ameliorates the clinical severity of MS and exhibits antiinflammatory, neuroprotective activities, and good tolerance for long-term use, whereas it is toxic to the CNS. Recently, the immunomodulation and neuroprotection capabilities of human bone marrow mesenchymal stem cells (hBM-MSCs) were shown in experimental autoimmune encephalomyelitis (EAE). In this study, we evaluated whether the combination of hBM-MSCs and a low-dose minocycline could produce beneficial effects in EAE mice.

Methods: The sensitivity of hBM-MSCs to minocycline was determined by an established cell-viability assay. Minocycline-treated hBM-MSCs were also characterized with flow cytometry by using MSC surface markers and analyzed for their multiple differentiation capacities. EAE was induced in C57BL/6 mice by using immunization with MOG35-55. Immunopathology assays were used to detect the inflammatory cells, demyelination, and neuroprotection. Interferon gamma (IFN-γ)/tumor necrosis factor alpha (TNF-α) and interleukin-4 (IL-4)/interleukin-10 (IL-10), the hallmark cytokines that direct Th1 and Th2 development, were detected with enzyme-linked immunosorbent assay (ELISA). terminal dUTP nick-end labeling (TUNEL) staining was performed to elucidate the cell apoptosis in the spinal cords of EAE mice.

Results: Minocycline did not affect the viability, surface phenotypes, or differentiation capacity of hBM-MSCs, while minocycline affected the viability of astrocytes at a high dose. In vivo efficacy experiments showed that combined treatment, compared to the use of minocycline or hBM-MSCs alone, resulted in a significant reduction in clinical scores, along with attenuation of inflammation, demyelination, and neurodegeneration. Moreover, the combined treatment with hBM-MSCs and minocycline enhanced the immunomodulatory effects, which suppressed proinflammatory cytokines (IFN-γ, TNF-α) and conversely increased anti-inflammatory cytokines (IL-4, IL-10). In addition, TUNEL staining also demonstrated a significant decrease of the number of apoptotic cells in the combined treatment compared with either treatment alone.

Conclusions: The combination of hBM-MSCs and minocycline provides a novel experimental protocol to enhance the therapeutic effects in MS.

Figure 1

Effects of minocycline on hBM-MSC stability. (A) hBM-MSC and astrocyte viabilities were analyzed with the MTT assay 24 hours after minocycline (0 to 10 μM) treatment. Minocycline did not affect hBM-MSC viability until 10 μM, whereas high concentrations (8 to 10 μM) decreased astrocyte viability. Points, mean; bars, SEM. (B) FACS analysis of the effect of minocycline on hBM-MSC phenotype. Wild-type (black lines) and minocycline-treated hBM-MSCs (red lines) were labeled with antibodies for MSC phenotypic surface markers. Minocycline did not change surface-marker expression; hBM-MSCs expressed CD90, CD44, and CD73 and lacked CD34, CD45, and HLA-DR. (C) The effect of minocycline on the differentiation potential of hBM-MSCs. Minocycline did not affect the differentiation capability of hBM-MSCs to adipogenic or osteogenic lineages, as stained by Oil Red O and Alizarin Red S, respectively. The results are representative of three independent experiments.

Figure 2

The combination of hBM-MSCs and minocycline attenuates clinical EAE severity in mice. EAE was induced in female C57 BL/6 mice by immunization with the MOG 35- to 55-amino acid peptide. (A) Mean daily clinical scores for the different EAE treatments (n = 10 per group) revealed that PBS-treated mice developed EAE with an average maximum severity over grade 4. Treatment with hBM-MSCs or minocycline alone decreased disease severity, with an average maximum severity of grades 1 to 2. Combination treatment markedly attenuated disease severity, with an average maximum severity of grade 1. Points, mean; bars, SEM. (B) The average clinical score of the four groups was assessed from day 1 until day 50 after immunization. Compared with PBS treatment, hBM-MSCs or minocycline treatment alone significantly decreased the average clinical scores (P < 0.05). Compared with hBM-MSCs or minocycline treatment alone, combination treatment significantly decreased the average clinical scores (P < 0.05). (C) The maximum clinical score for each mouse over the course of the entire experiment was recorded, and the trend of the maximum clinical scores in the four groups was the same as the average clinical score (P < 0.05). Columns, mean; bars, SEM. *P < 0.05, **P < 0.01, ***P < 0.001, one-way ANOVA with post hoc Bonferroni corrections. The results are representative of three independent experiments.

Figure 3

The combination of hBM-MSCs and minocycline decreases inflammation in EAE mouse spinal cord. (A) H&E (upper panel) and CD4 staining (bottom panel) were performed to detect mononuclear and T-cell infiltration in the lumbar spinal cords of EAE mice. The arrows indicate areas with infiltrated cells. Scale bar, 200 μm. (B) Statistical analysis showing that compared with PBS treatment, hBM-MSCs or minocycline treatment alone significantly decreased the average clinical scores (P < 0.05); compared with hBM-MSCs or minocycline treatment alone, combination treatment significantly decreased the numbers of infiltrating cells by 20% to 30% (P < 0.05). (C) Compared with hBM-MSCs or minocycline treatment alone, combination treatment significantly decreased the numbers of CD4⁺ cells by 30% to 40% (P < 0.05). Columns, mean; bars, SEM. *P < 0.05, **P < 0.01, ***P < 0.001, one-way ANOVA with post hoc Bonferroni corrections. The results are representative of three independent experiments.

Figure 4

The combination of hBM-MSCs and minocycline decreases demyelination in the EAE mouse spinal cord. (A) LFB- (upper panel) and MBP-stained (bottom panel) spinal cord sections were assessed to detect the extent of demyelination in EAE mice. The arrows indicate demyelinated areas. Scale bar, 200 μm. (B) Statistical analysis demonstrated that combination treatment significantly reduced demyelination compared with hBM-MSCs or minocycline treatment alone (P < 0.05). (C) Combination treatment also significantly preserved the number of MBP-positive cells compared with hBM-MSCs or minocycline treatment alone (P < 0.001). Columns, mean; bars, SEM. *P < 0.05, **P < 0.01, ***P < 0.001, one-way ANOVA with post hoc Bonferroni corrections. The results are representative of three independent experiments.

Figure 5

Combined treatment reduces glial reactivity and protects neurons in the EAE mouse spinal cord. (A) Sections are labeled for GFAP (upper panel), Iba-1 (middle panel), and NeuN (bottom panel) immunoreactivity to detect astrocytes, microglia, and neurons, respectively. Intense GFAP immunoreactivity was present in PBS-treated EAE mice and was reduced in EAE mice treated with hBM-MSCs or minocycline alone, but fewer astroglial cells were activated in the combination-treatment group. Iba-1, which identifies activated microglia, displayed the same immunoreactivity patterns as GFAP in the four EAE groups. The opposite pattern was observed for the neuronal marker, NeuN, in all groups. Scale bar, 200 μm. (B) Stereologic analyses of GFAP fluorescence intensity and (C) the number of Iba-1-positive cells revealed significant reductions in astrocytes and microglial activation in combination-treatment mice compared with those treated with hBM-MSCs or minocycline alone (P < 0.05). (D) Stereologic analysis also revealed a significant increase in the number of NeuN-positive cells in spinal cord sections of EAE mice treated with both hBM-MSCs and minocycline compared with hBM-MSCs or minocycline treatment alone (P < 0.01). Columns, mean; bars, SEM. *P < 0.05, **P < 0.01, ***P < 0.001, one-way ANOVA with post hoc Bonferroni corrections. The results are representative of three independent experiments.

Figure 6

Combination treatment promotes a shift from the Th1 to Th2 cytokine balance in EAE mice. Sera were isolated from the five groups of EAE mice 40 days after immunization. Cytokine protein-expression levels in the serum were quantified with ELISA. (A) Stereologic analysis demonstrated a significant decrease of IFN-γ/TNF-α Th1 cytokines in the hBM-MSCs or minocycline-treatment groups compared with PBS treatment, and a similar expression pattern was evident in the combination-treatment group compared with hBM-MSCs or minocycline treatment-alone groups. (B) Significant increases were present in IL-4 and IL-10 protein-expression levels in the hBM-MSCs or minocycline-treatment groups compared with PBS treatment, and the same pattern was evident in the combination-treatment mice compared with hBM-MSCs or minocycline treatment alone. Columns, mean; bars, SEM. *P < 0.05, **P < 0.01, ***P < 0.001, one-way ANOVA with post hoc Bonferroni corrections. The results are representative of three independent experiments.

Figure 7

hBM-MSCs and minocycline in combination reduce apoptosis in EAE mouse spinal cord. (A) Apoptotic cell death was examined with TUNEL staining. TUNEL-positive cells are red, and counterstaining for DAPI is blue. Scale bar, 200 μm. (B) TUNEL-positive cells were also quantified with MetaMorph image analysis. A statistically significant reduction in the number of apoptotic cells in the hBM-MSCs or minocycline treatment-alone group compared with the PBS-treated group (P < 0.001), and the combination treatment showed a more significant decrease in apoptosis compared to hBM-MSCs or minocycline treatment alone (P < 0.001). Columns, mean; bars, SEM. **P < 0.01, ***P < 0.001, one-way ANOVA with post hoc Bonferroni corrections. The results are representative of three independent experiments.