You are here

| Pluripotent Stem Cells

Immunogenicity of Induced Pluripotent Stem Cells

From Nature
By Stuart P. Atkinson

For many, induced pluripotent stem cell (iPSC) generation holds the key to the generation of patient-specific (autogenic) cells and tissues which could be used to treat various conditions and diseases. Therefore, one would expect that cells differentiated from iPSC which have been generated from our own somatic cells would be immune-tolerated. However, this assumption has never been tested. A recent proof (Zhao et al) published as an advanced online article on Nature from the lab of Yang Xu at the Division of Biological Sciences, University of California, San Diego now begins to address this concern.

Zhao et al first utilised embryonic stem cells (mESCs) from different strains of mice and also derived iPSCs using conventional retrovirally-mediated gene transduction (ViPSC) or episomal transduction to inhibit integration of transgenes (EiPSCs) from different strains of mice, to study the immunogenicity of transplanted cells. All iPSCs generated were deemed to be pluripotent which, importantly for this study, included the ability to form teratomas in SCID mice. Initial tests using mESC from C57BL/6 (B6) and 129/SvJ (129) strains showed that syngeneic transplantation of mESC (B6-ESC into B6 mice, and 129-ESC into 129 mice, sufficiently identical and immunologically compatible but not truly autologous) efficiently caused teratomas with no sign of immune rejection. However, when B6-mice were injected with 129-mESC (allogenic transplantation in so much that the transplantation occurs between genetically non-identical members of the same species) the cells were rejected and any teratomas which had formed showed massive amounts of infiltration by immunogenic cells (CD4+).

From this data, we would therefore expect a similar set of responses from syngeneic iPSCs when studying the immunogenic effects of autologous iPSC transplantation. However, when ViPSCs generated from B6 fibroblasts (B6-ViPSCs) were implanted into B6-mice, few teratomas were formed and those that were formed were massively infiltrated with T-cells and showed signs of necrosis and tumour regression. This suggests that syngeneic ViPSCs are highly immunogenic and are mainly rejected. There has been evidence in humans that there is a lack of immune tolerance to OCT4 and, indeed, OCT4-specific T cells can be readily detected in healthy humans (Dhodapkar et al) suggesting that reactivation of exogenous Oct4 expression in iPSCs could elicit an immune response. Therefore, to discount this hypothesis iPSC were generated using an episomaly-mediated gene transduction protocol (EiPSC) to ensure no integration of transgenes. However, similarly to results observed with ViPSC, teratomas formed upon syngeneic EiPSC implantation still demonstrated infiltration of T-cells, while regression and necrosis was detected in about 10% of cases. This suggests that while this immune response is lower than for the ViPSCs, an immunogenic response to allogenic EiPSCs still exists. The bottom line is that syngeneic iPSCs elicit an immune response leading to tumour regression and immune rejection, casting a doubt over their usefulness for regenerative medicine, assuming that this syngeneic transplantation is a sufficiently good model of the autologous manner in which iPSCs would be utilised.

Expression analysis comparing teratomas generated from B6-ESC or B6-EiPSC led to the discovery of 9 genes (Lce1f, Spt1, Cyp3a11, Zg16, Lce3a, Chi3l4, Olr1, Retn and Hormad1) which were over-expressed in regressing teratomas derived from EiPSCs. Of these genes, Hormad1 was identified as a tumour antigen and Spt1 as a tissue-specific antigen. To test the effect of these genes, 7 were ectopically over-expressed in B6-ESCs and the resulting teratomas studied. Eighty percent of Zg16-B6-ESC and 50% of Hormad1- and Cyp3a11-B6-iPSCs implants failed to form any teratomas, with T-cell infiltration and necrosis observed in those that had formed. Further studies identified that CD4+ helper T-cells and CD8+ cytotoxic T-cells were important for this immune rejection, as cells lacking CD4 and CD8 cells did not show any immune response. It was also discovered that Zg16 and Hormad1 had a direct effect on T-cells, as forced expression induced IFNg release from purified T-cells, indicative of an immune response dictated by these genes.

Overall, these data suggest that syngeneic iPSCs elicit a significant immune response following transplantation. The authors suggest that this could be due to a response to “abnormal” antigens expressed by the iPSCs which do not appear during normal development and differentiation, such as OCT4, perhaps fuelled by epigenetic and genetic differences observed in iPSC compared with ESC. How much of an impact could this data have? Firstly, this data pertains to implanted pluripotent cells, whereas implantation of cells for tissue replacement would likely use differentiated cell types and therefore the immunogenic response to differentiated syngeneic tissues must also be studied. Secondly, the said experiments may not mimic the human situation accurately, as it is not described if the tail tip fibroblast-derived iPSCs were transplanted back into the same mouse from where the donor cells originated an important consideration. The experiments described are thus syngeneic rather than autologous. Thirdly, this data does suggest that perhaps we still need to refine the reprogramming process in order to minimise the significant differences which remain between iPSC and ESC. This perhaps arises during the reprogramming process but also the growth of ESC/iPSCs in vitro might imbue them with characteristics that are refractory for implantation, possibly arising from the growth substrate, growth medium and general culture/passage conditions during their expansion.



Immunogenicity of induced pluripotent stem cells
Tongbiao Zhao, Zhen-Ning Zhang, Zhili Rong & Yang Xu
Nature (2011) doi:10.1038/nature10135
Received 07 July 2010 Accepted 19 April 2011 Published online 13 May 2011

Natural immunity to pluripotency antigen OCT4 in humans.
Dhodapkar KM, Feldman D, Matthews P, Radfar S, Pickering R, Turkula S, Zebroski H, Dhodapkar MV.
Proc Natl Acad Sci U S A. 2010 May 11;107(19):8718-23. Epub 2010 Apr 19.