A 3D sphere culture system containing functional polymers for large-scale human pluripotent stem cell production

Abstract

Utilizing human pluripotent stem cells (hPSCs) in cell-based therapy and drug discovery requires large-scale cell production. However, scaling up conventional adherent cultures presents challenges of maintaining a uniform high quality at low cost. In this regard, suspension cultures are a viable alternative, because they are scalable and do not require adhesion surfaces. 3D culture systems such as bioreactors can be exploited for large-scale production. However, the limitations of current suspension culture methods include spontaneous fusion between cell aggregates and suboptimal passaging methods by dissociation and reaggregation. 3D culture systems that dynamically stir carrier beads or cell aggregates should be refined to reduce shearing forces that damage hPSCs. Here, we report a simple 3D sphere culture system that incorporates mechanical passaging and functional polymers. This setup resolves major problems associated with suspension culture methods and dynamic stirring systems and may be optimal for applications involving large-scale hPSC production.

Graphical abstract

Figure 1

Sphere Culture of Human Pluripotent Stem Cells (A) Mechanical subculture by passaging the cells through mesh filters. Spheres are pushed through a mesh filter with an opening size of 50 μm using a 1 ml micropipette tip. The scale bar represents 100 μm. (B) Subculture cycle and shapes of spheres of the KhES-1 cell line during the 5-day interval. The scale bar represents 200 μm. (C) Size distribution and morphology of cells from the 253G1 cell line in culture media with or without 0.3% methylcellulose on day 5. The blue and red bars represent the number of spherical and fused spheres in each size range, respectively. The graphs show exemplary one of the five independent experiments. The scale bars represent 200 μm. (D) Comparison of expansion rates of the KhES-1, H9, and 253G1 cell lines in the sphere culture or in the conventional adherent culture on feeder layers or on Matrigel. The graph shows exemplary expansion rate plotting. Average slopes and SD in the semilogarithmical plotting were obtained by calculating exponential trend lines from three independent experiments and indicated in each graph. See also Figures S1, S2, and S4.

Figure 2

Transmission Electron Microscopy of Human Pluripotent Stem Cell Spheres (A) A low-magnification (light microscopy) image of a KhES-1 sphere at four passages. The blue signal indicates nuclei. The scale bar represents 50 μm. (B) A high-magnification image of the peripheral region of a KhES-1 sphere. The black arrows indicate adherens junctions, and the red signal indicates actin bundles. The black dots indicate glycogen granules. The scale bar represents 1 μm. (C–F) Higher magnification images of a (C) tight junction, (D) adherens junction, (E) desmosome, and (F) gap junction found in the periphery of a KhES-1 sphere. The scale bar represents 100 nm.

Figure 3

Pluripotency Marker Expression and Karyotype of Human Pluripotent Stem Cells in Sphere Culture (A) Immunofluorescent staining of pluripotency marker proteins NANOG, OCT3/4, SSEA-4, and TRA-1-60 in frozen sections of KhES-1 spheres after 55 passages. In the merged panels, the green signals indicate the markers and the blue signals indicate DAPI nuclear staining. The scale bar represents 100 μm. (B) Flow cytometry analysis of the pluripotency surface markers SSEA-4, TRA-1-60, and SSEA-3 on KhES-1 cells in the sphere culture after 72 passages. The control panel indicates analysis without the primary antibody. The percentage of the marker-positive population is indicated in each panel. (C) Multicolor FISH karyotype analysis of sphere-cultured KhES-1 cells at passage 51. (D) Histology of teratomas derived from KhES-1 cells at passage 21 in the sphere culture. The teratomas contained various tissues from the three germ layers, including the neural epithelium (ectoderm), cartilage (mesoderm), and gut-like epithelium with mucosa (endoderm). The scale bars represent 50 μm. See also Figures S3–S5.

Figure 4

Characterizations of Low-Acyl Gellan Gum Polymer (A) The chemical structure of the repeat unit of low-acyl gellan gum (GG). (B) Stereo view of GG (reproduced from Figure 2 in Chandrasekaran and Thailambal, 1990). Two adjacent up- and down-pointing gellan double-helices are crosslinked at the arrows by calcium ions (filled circles). (C) Apparent viscosities and settling rates of GG and methylcellulose (MC). The average and SD are shown from three experiments (n = 3). The asterisks indicate no settling. (D) Settled or suspended polystyrene beads in the culture medium under various concentrations of GG.

Figure 5

3D Sphere Culture of Human Pluripotent Stem Cells (A) Complete inhibition of hPSC sphere sedimentation by low-acyl GG at 0.015%. KhES-1 spheres on day 4 were suspended in the culture medium with various concentrations of GG and observed after 16 hr. The scale bar represents 5 mm. (B) Fold increase in KhES-1 cell number in culture medium with or without GG in different culture vessels. The average fold increase from independent experiments indicates the cell number increase from days 0 to 5, and the error bars indicate the SD (n = 10 for tube-shaped bags and n = 4 for others). Significance calculations were performed using the Student’s t test. (C) A tube-shaped culturing bag made of a gas-permeable membrane (left) and polystyrene tubes (15 and 5 ml). (D) Morphologies of KhES-1 spheres on day 5 in the 3D sphere culture medium with 0.020% GG. The scale bar represents 200 μm. (E) Comparison of KhES-1 cell expansion rates in the 3D sphere culture using tube-shaped gas-permeable bags or culture plates. The graph shows exemplary expansion rate plotting. Average slopes and SD in the semilogarithmical plotting were obtained by calculating exponential trend lines from three independent experiments and indicated in each graph. See also Figure S6 and Movie S1.

Figure 6

Proof of Principle 3D hPSC Sphere Culture (A) A trial for larger-scale sphere culture by using 200 ml gas-permeable membrane bags. The scale bar represents 5 cm. (B) Comparison of cell yield calculated from the average cell number obtained in the 3D sphere or adherent culture using KhES-1 cell line. (C) Morphology of KhES-1 spheres cultured for 5 days in a 200 ml bag. The scale bar represents 200 μm.