When mouse embryonic stem cells were first derived in 1981 independently by Gail Martin at UCSF and Evans and Kaufman at Cambridge University, the inner cell mass cells from the blastocyst-stage mouse embryos were cultured on a layer of mouse skin cells that had been treated with a drug that prevented them from dividing or with radiation that did the same. These single layers of mouse skin fibroblasts secreted growth factors that prevented the embryonic stem cells from differentiating and drive them to divide. These layers of cells were known as “feeder” cells, because the secretions of the cells fed the growing embryonic stem cells.
When James Thomson at the University of Wisconsin derived the first human embryonic stem cell lines in 1998, he also used mouse feeder cells to keep the cells growing and undifferentiated. Once the embryonic stem cells were taken from this culture system, they began to differentiate.
However, it became equally clear that using mouse feeder cells represented a problem if human embryonic stem cells were going to be used for clinical purposes because animal cells can harbor occult viruses and other infectious agents that can infect human cells. Also, animal cells possess unusual sugars that are transferred to human cells when they are together in culture. Such foreign sugars can elicit robust immune responses against the cells if they are used for clinical purposes See Martin et al., Nature Medicine 2005; 11:228-232; and Stacey et al., Journal of Biotechnology 2006;125:583-588). Therefore, it became clear that finding ways to grow embryonic stem cells in the absence of feeder lines was an important goal if these cells were going to be used for clinical purposes.
Several laboratories successfully derived so-called “Xeno-free” embryonic stem cells by using protein substrata to grow the cells. These protein substrata included matrigel (animal), human laminin, E-cadherin, and vitronectin (see Xu C,, et al (2001) Nat Biotechnol 19:971–974; Miyazaki T,, et al. (2008) Biochem Biophys Res Commun 375:27–32; Nagaoka M,, et al. (2010) BMC Dev Biol 10:60; Chen G,, et al. (2011) Nat Methods 8:424–429). When Yamanaka and his colleagues discovered procedures for making human induced pluripotent stem cells, once again, feeder lines were initially used, but feeder-free protocols were also developed for deriving xeno-free induced pluripotent stem cells (iPSCs; see Chen G,, et al. (2011) Nat Methods 8:424–429; Nakagawa M,, et al. (2014) Sci Rep 4:3594).
A new report from Luis Gerardo Villa-Diaz, Jin Koo Kim, Joerg Lahann, and Paul H. Krebsbach from the University of Michigan, Ann Arbor, Michigan, has described a way to derive and grow human iPSCs on a completely synthetic substratum. This substratum, poly2-(methacryloxy)ethyl dimethyl-(3-sulfopropyl) ammonium hydroxide, or PMEDSAH, forms a hydrogel that is completely synthetic. Therefore, the cells do not touch anything made from genetically manipulated cells or animal products.
Krebsbach and his group used fibroblasts from human gum tissue as their cell source. These cells were reprogrammed into iPSCs by means of infection with recombinant Sendai viruses. These viruses cause expression of the four genes required for reprogram cells (Oct4, Klf4, Sox2, and c-Myc), but they do not insert their viral genomes into the chromosomes of the host cell. Therefore, these viruses only express the reprogramming factors transiently, and afterwards, no trace of them can be found in the iPSC line, provided you properly screen for the absence of the virus. The reprogrammed cells were grown on the PMEDSAH and the cells not only were reprogrammed on this substratum, but also grew on it rather well.
The gum-based fibroblasts were nicely reprogrammed and made iPSCs that expressed all the right genes and produced tumors called teratomas when implanted into nude mice. The teratoma-production assay is an important test for pluripotency, because teratomas are tumors that consist of the mishmash of different tissue types. The fact that implanted cells produce these tumors with a mixed cell population of such wildly different cell types is an important indication of their pluripotency.
Even more importantly perhaps is the genetic integrity of these cells. Karyotypes of these iPSC lines (karyotypes lay our the chromosomes of the cell to see if there are a normal number of chromosomes and if the chromosomes appear normal) revealed that they were beautifully normal. However, hCGH (array-based comparative genomic hybridization) analysis, which uses specific chromosome-specific probes to finding missing or duplicated bits of chromosomes that are too small to see in a karyotype revealed a few missing and added bits to the genomes of all three derived iPSC lines. None of these were in known cancer loci. As shown in the figure below, these lines had only a few mutations. The karyotypes used cells from nine months after their derivation. Thus these cells proved to be rather stable.
This paper demonstrates that it is possible to generate transgene-free, stable iPSCs on a synthetic substrate. This type of platform has the potential to meet the good manufacturing practices that must be used to make products for clinical use.