Pluripotent Nontumorigenic Adipose Tissue-Derived Muse Cells have Immunomodulatory Capacity Mediated by Transforming Growth Factor-β1

Abstract

Adult mesenchymal stromal cell-based interventions have shown promising results in a broad range of diseases. However, their use has faced limited effectiveness owing to the low survival rates and susceptibility to environmental stress on transplantation. We describe the cellular and molecular characteristics of multilineage-differentiating stress-enduring (Muse) cells derived from adipose tissue (AT), a subpopulation of pluripotent stem cells isolated from human lipoaspirates. Muse-AT cells were efficiently obtained using a simple, fast, and affordable procedure, avoiding cell sorting and genetic manipulation methods. Muse-AT cells isolated under severe cellular stress, expressed pluripotency stem cell markers and spontaneously differentiated into the three germ lineages. Muse-AT cells grown as spheroids have a limited proliferation rate, a diameter of ∼15 µm, and ultrastructural organization similar to that of embryonic stem cells. Muse-AT cells evidenced high stage-specific embryonic antigen-3 (SSEA-3) expression (∼60% of cells) after 7-10 days growing in suspension and did not form teratomas when injected into immunodeficient mice. SSEA-3+ -Muse-AT cells expressed CD105, CD29, CD73, human leukocyte antigen (HLA) class I, CD44, and CD90 and low levels of HLA class II, CD45, and CD34. Using lipopolysaccharide-stimulated macrophages and antigen-challenged T-cell assays, we have shown that Muse-AT cells have anti-inflammatory activities downregulating the secretion of proinflammatory cytokines, such as interferon-γ and tumor necrosis factor-α. Muse-AT cells spontaneously gained transforming growth factor-β1 expression that, in a phosphorylated SMAD2-dependent manner, might prove pivotal in their observed immunoregulatory activity through decreased expression of T-box transcription factor in T cells. Collectively, the present study has demonstrated the feasibility and efficiency of obtaining Muse-AT cells that can potentially be harnessed as immunoregulators to treat immune-related disorders. Stem Cells Translational Medicine 2017;6:161-173.

Keywords: Antigen-specific response; Immunomodulation; Spheroids/clusters; Stem cells; T lymphocytes.

© 2016 The Authors Stem Cells Translational Medicine published by Wiley Periodicals, Inc. on behalf of AlphaMed Press.

Figures

Figure 1

Generation and culture of liposuction‐derived Muse‐AT cells. (A): Scheme of Muse‐AT cell preparation protocol. Muse‐AT cells were obtained after digestion with collagenase and severe stress conditions. Clusters were generated when Muse‐AT cells were seeded on nonadherent plastic. (B): Muse‐AT cells formed clusters of 50–150 µm in diameter that grew in suspension culture. Scale bar = 50 µm. Left inset, black arrow indicates a single Muse‐AT cell at the edge of a cluster. Scale bar = 20 µm. (C): Transmission electron microscopy of a representative Muse‐AT cell after 5–7 days in culture showing a high nucleus to cytoplasm ratio. Scale bar = 500 nm. (D): Muse‐AT cells were seeded immediately on nonadhesive dishes after isolation (day 0) and formed clusters after 3 days of culture (day 3, cycle 1). After mechanical disaggregation, spheroids formed again, reaching 50–150 µm in diameter during the second and third growth cycle. Abbreviations: FBS, fetal bovine serum; Muse‐AT, multilineage‐differentiating stress‐enduring cells derived from adipose tissue; O/N, overnight; w/o, without.

Figure 2

Expression of pluripotency stem cell markers of Muse‐AT cell clusters. (A): Representative immunostaining of the stem cell markers Nanong, OCT4, TRA1‐60, Sox‐2, and SSEA‐4 were observed in almost all Muse‐AT cells in the clusters. Scale bar = 100 µm. White arrowheads indicate single cells shown in upper right insets. Scale bar = 5 µm. (B): Immunofluorescence quantification of stem cell markers. (C): SSEA‐3 was most highly expressed by Muse‐AT cells compared with ASCs, as assessed by fluorescence activated cell sorter (FACS) staining. (D): Immunofluorescence staining confirmed SSEA‐3 expression by Muse‐AT cell clusters. Scale bar = 100 µm. (E): FACS staining of several CDs revealed the immunophenotype of Muse‐AT cells. *, p < .05; **, p < .005; n = 3 samples analyzed. Abbreviations: ASCs, adipose‐derived stromal cells; d, day; Muse‐AT, multilineage‐differentiating stress‐enduring cells derived from adipose tissue; SSEA, stage‐specific embryonic antigen.

Figure 3

Muse‐AT cells differentiate into the three germ cell lineages. (A): Muse‐AT cells cultured in adherent plastic have the ability to spontaneously differentiate into mesodermal, endodermal, and ectodermal lineages, as demonstrated by the expression of mRNAs for Nkx2.5, α‐fetoprotein, and MAP‐2 respectively, determined by qualitative real‐time polymerase chain reaction. (B): In the presence of specific differentiation medium, immunofluorescence staining revealed positivity for hepatocyte (α‐fetoprotein, red), myocyte (SMA, green), and neuron (MAP‐2, red) markers. Nuclei were stained with 4′,6‐diamidino‐2‐phenylindole (blue). Original magnification ×200. Abbreviations: GAPDH, glyceraldehyde‐3‐phosphate dehydrogenase; MAP‐2, microtubule‐associated protein 2; Muse‐AT, multilineage‐differentiating stress‐enduring cells derived from adipose tissue; Nkx2.5, NK2 homeobox 5; SMA, smooth muscle actin.

Figure 4

Lack of teratoma formation after transplantation and stability of Muse‐AT cells. (A): Muse‐AT cells were injected (106) i.t. into NODscid mice. The transplanted mice were monitored weekly for the appearance of tumors. The P19 embryonic carcinoma cell line was injected (106) as the control. The mice were sacrificed when the tumors became outwardly apparent. NODscid mice injected with Muse‐AT cells did not develop teratoma during the observed period (up to 6 months). (B): H&E staining of testes showed normal tissue structure in NODscid Muse‐AT cells of the injected mice. (C): Representative normal karyotype of Muse‐AT cells that showed expression of stem cell markers. Abbreviations: d, day; i.t., intratesticular; Muse‐AT, multilineage‐differentiating stress‐enduring cells derived from adipose tissue.

Figure 5

Spontaneous expression of TGF‐β1 by Muse‐AT clusters. (A): Immunofluorescence staining of TGF‐β1 at different time points in culture. (B): Immunofluorescence quantification showed maximal expression of TGF‐β1 after 10 days in culture. (C): Expression levels of TGF‐β1 mRNA of Muse‐AT cell clusters at 0, 5, and 10 days in culture were determined by real‐time polymerase chain reaction. Ct values were normalized to the expression of the HPRT gene, and data are expressed relative to the values obtained for day 0. *, p < .05; **, p < .005; n = 5. Abbreviations: ASC, adipose‐derived stromal cell; d, day; Muse‐AT, multilineage‐differentiating stress‐enduring cells derived from adipose tissue; TGF‐β1, transforming growth factor‐β1.

Figure 6

Muse‐AT cell activity on LPS‐stimulated macrophages and splenocytes. (A): Muse‐AT cells in coculture with the RAW macrophage cell line. RAW cells were seeded onto the lower chamber of a Transwell plate (Corning) and stimulated with LPS for 20 minutes. LPS was removed, the cells washed, and Muse‐AT cells were then cultured (RAW/Muse‐AT in a 1:10; 1:1, and 10:1 ratio) in the upper chamber. (B): Muse‐AT cell CM effects on LPS‐stimulated RAW cells with and without the presence of SB 431542 (SB; 10 ng/ml). (C, D): Freshly isolated murine peritoneal MΦ stimulated with LPS. The effect of Muse‐AT cell CM on the secretion of TNF‐α (C) and IL‐10 (D) was evaluated with and without SB in the culture medium. (E): Antigen‐specific T‐cell response of NOD BDC2.5 splenocytes in culture. Muse‐AT cell CM was assayed at different dilutions (1 to 5 to 1 to 5,000; n = 3). (F): Muse‐AT cell CM downregulated antigen‐stimulated T‐bet expression assessed by Western blot (n = 3). (G): Inhibitory effect on antigen‐challenged T cells on TNF‐α secretion by Muse‐AT cell CM (n = 3). (H): Muse‐AT cell CM enhanced anti‐inflammatory cytokine IL‐10 secretion in antigen‐stimulated T cells (n = 6). The expression levels of mRNA IL‐10 relative to mRNA HPRT are indicated as an inset. Cytokine levels were quantified in 72‐hour culture supernatants by enzyme‐linked immunosorbent assays. OVA was used as a control. Mi concentration = 5 nM. *, p < .05; **, p < .005; ***, p < .0005. Abbreviations: ASCsCM, adipose stem cell conditioned media; CM, conditioned media; GAPDH, glyceraldehyde‐3‐phosphate dehydrogenase; IL, interleukin; LPS, lipopolysaccharide; MΦ, macrophages; Mi, mimotope; Muse‐AT, multilineage‐differentiating stress‐enduring cells derived from adipose tissue; NOD, nonobese diabetic; OVA, ovalbumin; Spl, splenocytes; T‐bet, T‐box transcription factor (TBX21); TNF‐α, tumor necrosis factor‐α.

Figure 7

TGF‐β1 signaling blockade on cytokine secretion. (A): Exogenously added TGF‐β1 to antigen‐stimulated splenocytes showed a concentration response dependence on IFN‐γ secretion. (B): The effect of TGF‐β1 signaling blockade was assessed using SB at different concentrations (1.25–10 ng/ml) on antigen‐challenged T cells followed IL‐10 quantification. (C): Inhibition of TGF‐β1 signaling with 10 ng/ml of SB restored IFN‐γ secretion levels. (D): SB diminished pSMAD2 levels assessed by Western blot (WB) in antigen‐stimulated T cells. (E): Quantification of WB analysis shown in (D). Cytokine measured by enzyme‐linked immunosorbent assays in 72‐hour supernatants. *, p < .05; **, p < .005; ***, p < .0005; n = 3. Abbreviations: CM, conditioned media; GAPDH, glyceraldehyde‐3‐phosphate dehydrogenase; IFN‐γ, interferon‐γ; Mi, mimotope; Muse‐AT, multilineage‐differentiating stress‐enduring cells derived from adipose tissue; p, phosphorylated; SB, SB 431542; Spl, splenocytes; TGF‐β1, transforming growth factor‐β1.