Selective development of myogenic mesenchymal cells from human embryonic and induced pluripotent stem cells

Abstract

Human embryonic stem (ES) cells and induced pluripotent stem (iPS) cells are promising sources for the cell therapy of muscle diseases and can serve as powerful experimental tools for skeletal muscle research, provided an effective method to induce skeletal muscle cells is established. However, the current methods for myogenic differentiation from human ES cells are still inefficient for clinical use, while myogenic differentiation from human iPS cells remains to be accomplished. Here, we aimed to establish a practical differentiation method to induce skeletal myogenesis from both human ES and iPS cells. To accomplish this goal, we developed a novel stepwise culture method for the selective expansion of mesenchymal cells from cell aggregations called embryoid bodies. These mesenchymal cells, which were obtained by dissociation and re-cultivation of embryoid bodies, uniformly expressed CD56 and the mesenchymal markers CD73, CD105, CD166, and CD29, and finally differentiated into mature myotubes in vitro. Furthermore, these myogenic mesenchymal cells exhibited stable long-term engraftment in injured muscles of immunodeficient mice in vivo and were reactivated upon subsequent muscle damage, increasing in number to reconstruct damaged muscles. Our simple differentiation system facilitates further utilization of ES and iPS cells in both developmental and pathological muscle research and in serving as a practical donor source for cell therapy of muscle diseases.

Figure 1. Schematic presentation of the differentiation protocol.

Two different differentiation protocols were compared (Left: embryoid body (EB) culture, Right: Dissociation culture) and were exactly the same until the first 21 (7+14) days of culture. On the left side, EBs continued to be incubated in serum-containing medium without specific manipulation until the end of culture. On the right side, EBs and their outgrowth cells were dissociated and seeded onto collagen type I-coated tissue culture plates in serum-containing medium. The medium was changed to serum-free ITS medium on day 49 (7+14+28). In some experiments, the cells were harvested and used as donor cells for the transplantation assay at this time point.

Figure 2. Skeletal muscle development from human embryonic stem (hES) and human induced pluripotent stem (hiPS) cells by the EB culture method.

(A) PAX3- and PAX7-positive nuclei emerged in the proximal area of the embryoid body (EB)-outgrowth cells derived from hES KhES1 cells. (B) Simultaneous derivation of neural and cardiac cells in the EB-outgrowth cells derived from hES KhES1 cells. Upper: TUJ1-positive neural cells observed on day 7+28. Lower: Neural cells (outlined arrowheads) and colonies of beating cardiomyocytes (white arrowhead) appeared on day 7+28. (C) Skeletal myosin-positive myofibers in the EB-outgrowth cells derived from human embryonic stem (hES) KhES1 cells detected on day 7+42. (D) Sequential analysis of undifferentiated and skeletal myogenesis-related gene expression by semi-quantitative RT-PCR. (E) Skeletal myosin-positive fibers from human induced pluripotent stem (hiPS) cells. Four hiPS cell-lines were used. hiPS 201B6 on day 7+105, hiPS 201B7 on day 7+105, hiPS 253G1 on day 7+77, and hiPS 253G4 on day 7+56. (F) Sequential analysis of undifferentiated and skeletal myogenesis-related gene expression by semi-quantitative RT-PCR. In (A-C) and (E), antibodies were visualized using Cy3 (red). Nuclei were counterstained with DAPI (blue). Scale bars  = 100 µm.

Figure 3. Characterization and differentiation of the derived myogenic mesenchymal cells.

(A) Morphology of the derived myogenic mesenchymal progenitors 2 days (day 7+14+2) and 4 weeks (day 7+14+28) after replating. Homogeneous spindle-shaped fibroblastic cells were observed. (B) Surface marker analysis of myogenic mesenchymal progenitors. Representative data from KhES1 differentiation are shown. Note that CD56 in addition to mesenchymal markers CD73, CD105, CD166, and CD29 was exclusively expressed. (C) Changes in the expression of myogenic markers were analyzed by immunofluorescence. The number of Cy3-positive nuclei was divided by the total number of nuclei stained by DAPI. The expression of myogenic progenitor markers decreased after exposure to serum-free medium, whereas the number of MYOG-positive cells substantially increased after serum deprivation. (D) Changes in the number of MYOG-positive nuclei were observed up to 3 weeks after serum deprivation. hES/iPS-derived myofibers tended to detach from tissue culture plates during long-term culture in serum-free medium. (E) Serum deprivation increased the number of skeletal myosin-positive fibers and MYOG-positive nuclei for more than 2 weeks. KhES1 was used in this figure. (F) Multinucleated myofibers denoted by MYOG myogenin-positive nuclei aligned in skeletal myosin-positive fibers. (G) Morphology of mature myofibers, which were stained with skeletal myosin, MYOG, and dystrophin, from both KhES1 and 253G4 cells. Skeletal myosin was visualized with fluorescein isothiocyanate (FITC) (Green), myogenin was visualized with Cy3 (red), and nuclei were counterstained with DAPI (blue). Scale bars  =  (C, E) 100 µm, (D) 50 µm.

Figure 4. Engraftment of myogenic progenitors in damaged muscles of immunodeficient mice.

(A) Human nuclei labeled with human-specific lamin A/C localized mainly inside muscle fibers surrounded by laminin. (B) Muscle reconstruction by transplanted human cells was demonstrated by the detection of human-specific laminin-alpha 2. (C) The proportion of myofibers containing human nuclei at 4, 12, and 24 weeks after transplantation. (D) The proportion of myofibers containing human nuclei in reinjured (3+1 weeks) and in non-reinjured mice (4 weeks) at 4 weeks after transplantation. In C and D, data are presented as the mean ± standard deviation. (E) Distribution of the transplanted cells at 24 weeks after transplantation. Typical central nuclei of human origin were observed (outlined arrowheads). Some human cells located within the lamina rara beneath the basal lamina, indicating engraftment of the transplanted cells into a satellite cell compartment (white arrowhead). (F) Triple-staining for human Lamin A/C, PAX7, and pan-Laminin clearly demonstrated the existence of PAX7-positive human nuclei indicating the transplanted cells engrafted as satellite cells (white arrowhead). Human lamin A/C-negative host satellite cells were also detected (outlined arrowhead). Laminin was stained by a polyclonal antibody that recognizes both human and murine laminin, and was subsequently visualized with fluorescein isothiocyanate (FITC) (Green); human lamin A/C and human-specific laminin, with Cy3 (red). Nuclei were counterstained with DAPI (blue). Scale bars  =  (A) 100 µm, (B) and (E) 50 µm.