Generation of keratinocytes from normal and recessive dystrophic epidermolysis bullosa-induced pluripotent stem cells

Abstract

Embryonic stem cells (ESCs) have an unlimited proliferative capacity and extensive differentiation capability. They are an alternative source for regenerative therapies with a potential role in the treatment of several human diseases. The clinical use of ESCs, however, has significant ethical and biological obstacles related to their derivation from embryos and potential for immunological rejection, respectively. These disadvantages can be circumvented by the alternative use of induced pluripotent stem cells (iPSCs), which are generated from an individual's (autologous) somatic cells by exogenous expression of defined transcription factors and have biological characteristics similar to ESCs. In recent years, patient-specific iPSCs have been generated to study disease mechanisms and develop iPSC-based therapies. The development of iPSC-based therapies for skin diseases requires successful differentiation of iPSCs into cellular components of the skin, including epidermal keratinocytes. Here, we succeeded in generating iPSCs not only from normal human fibroblasts but also from fibroblasts isolated from the skin of two patients with recessive dystrophic epidermolysis bullosa. Moreover, we differentiated both of these iPSCs into keratinocytes with high efficiency, and generated 3D skin equivalents using iPSC-derived keratinocytes, suggesting that they were fully functional. Our studies indicate that autologous iPSCs have the potential to provide a source of cells for regenerative therapies for specific skin diseases.

Fig. 1.

Generation of normal human iPSCs (A–H) and patient-specific iPSCs (I–P) from a representative patient with RDEB. We show the morphology of hiPS1 (A and B) and hiPS-RDEB1 (I and J), and immunocytochemical analysis of stem cell markers, alkaline phosphatase (AP; C and K), OCT4 (D and L), SOX2 (E and M), SSEA3 (F and N), TRA-1–61 (G and O), and TRA-1–80 (H and P) in these iPSCs. iPSCs had morphology and gene expression patterns similar to ESCs. (A and I) Low magnification in bright field. (B and J) High magnification in bright field.

Fig. 2.

Characteristics of iPSCs. (A) qPCR analysis of the four endogenous (h) and four viral (v) factors, c-MYC, SOX2, OCT4, and KLF4. The four endogenous factors were expressed, and viral transgene silencing was observed in generated iPSCs. Black, endogenous expression of four factors. Gray, viral transgene expression of four factors. (B) RT-PCR analysis of stem cell markers in normal and PS-iPSCs. (C) Karyotype analysis. Both iPSCs had normal male karyotypes (46, XY) consistent with the starting fibroblasts. (D) Methylation analysis of NANOG promoter region in iPSCs. Unmethylated (○) and methylated (●) CpGs. NANOG promoter region in iPSCs were almost completely unmethylated, similar to ESCs.

Fig. 3.

Differentiation analyses of iPSCs. We confirmed differentiation capabilities of hiPS1 and hiPS-RDEB1 into all three germ layers by embryoid body formation in vitro and teratoma formation in vivo. Beta-III tubulin (TUBIII), neuron, and neural epithelium indicate ectodermal differentiation. Vimentin (VIM) and cartilage indicate mesoderm differentiation. α-fetoprotein (AFP) and gut epithelium indicate endoderm differentiation.

Fig. 4.

Directed differentiation of iPSC into keratinocytes in vitro. (A) Schematic representation of differentiation strategy for generating keratinocytes from iPSCs. RA and BMP4 were added on day 0 and 2 (black arrowheads). Cells were passaged for the first time at day 30 (white arrowhead). (B) Morphology of iPSC-derived cells (passage 4). A large number of epithelial cells were obtained under our differentiation conditions. (C) Differentiation efficiency of iPSC-derived K14+ cells on day 30 shows that the efficiency for inducing K14+ cells was highest with use of defined keratinocyte serum-free medium (KSFM). (D) Differentiation efficiency of iPSC-derived K14+ cells after first passage shows the possibility that culture conditions may enrich K14+ cells.

Fig. 5.

Characteristics and differentiation capacity of iPSC-derived keratinocytes. Both keratinocytes derived from normal (E–H) and PS-iPSCs (I–L) expressed keratinocyte markers, including keratin 14 (KRT14), p63 (A, E, and I) and desmoglein 3 (DSG3; B, F, and J), similar to normal human keratinocytes (nhKC; A–C). Type VII collagen (COL7A1) was only expressed by normal iPSC-derived keratinocytes (G), and not by PS-iPSC–derived keratinocytes (K) as expected from RDEB patients. Under high-calcium conditions, iPSC-derived keratinocytes expressed keratin 1, indicative of epidermal differentiation (H and L).

Fig. 6.

3D skin equivalents using iPSC-derived keratinocytes. 3D skin equivalents were generated using not only hiPSC but also hiPS-RDEB. The multilayered epidermis expressed keratin 1, laminin 5, and loricrin in 3D skin equivalents using normal keratinocytes and hiPS, demonstrating that iPSC-derived keratinocytes can be terminally differentiated in these skin equivalents. As expected, RDEB patient-specific keratinocytes failed to express type VII collagen. Clefting and parakeratosis are nonspecific findings observed in 3D skin equivalents. The gray line represents the dermal-epidermal boundary. E, epidermis; D, dermis.