New turn in iPS cell immunogenicity story

A new study, published today in Nature, challenges the previous conclusions about immunogenicity of iPS cells. You can read a great coverage of this story here, here and here. There are some contradictions between these studies, but is it really a defeat of Zhao’s study?

The contradictions:

• Zhao found that iPS-derived teratomas caused immune response, unlike ES cell-derived, but Akari did not observe immune response to either iPS- or ES cell-derived teratomas in syngenic mice.
• Zhao linked the immunogenicity to increased expression of Zg16 and Hormad1 genes, but Akari didn’t detect overexpression of these genes.
• Zhao concluded that iPS cell have significant immunogenicity, but Akari concluded that immunogenicity of iPS cells is negligible.

The main points of critical comments about Zhao’s study were: (1) the teratomas should cause immune response (see Yamanaka’s comment 1, comment 2) and (2) immunogenicity of differentiated tissues, derived from pluripotent cells, should be assessed.

If Yamanaka said that T-cell infiltration should be observed around teratomas, I wonder how he would comment Akari’s findings:

We did not observe any T cells in most sections, although fewer than two T cells were detected in some sections. Although slight differences were observed among the cell lines, the profiles between iPSCs and ES cells were quite similar overall…

So, it seem like Zhao’s experiment on immune response to iPS and ES cell-derived teratomas was not reproducible in Akari’s hands. By the way, I’ve heard on one of conferences last year, that Zhao’s experiment has been reproduced.

Experimental approach and conclusions:
Unlike Zhao’s study, Akari attempted to answer a question of immunogenicity of differentiated cells, derived from iPS cells. To do so, he used very sophisticated approach – generation of chimeric mice from GFP-labeled iPS cell lines and transplantation of “mature tissues” (skin and bone marrow) into syngenic donor. The immune response to these “iPS-matured cells” was negligible. Importantly, they found that “immunogenicity of iPSC-derived tissues is indistinguishable from that of ES-cell-derived tissues.” They attributed some degree of T-cell infiltration to inter-iPS cell lines variability rather than to difference between iPS and ES cell immunogenicity.

Some flaws and questions:
1. I wonder why pretty much the same experimental set up didn’t elicit any immune response to teratomas in Akari’s study:

Given that teratomas are a type of tumour, it is not surprising that they elicit immunological reactions in transplant recipients.

1 sentence below:

Hence, although we did not detect any significant T-cell infiltrations in various teratomas derived from either iPSCs or ES cells in our assay, we cannot rule out the possibility that substantial immune responses can be elicited under certain situations during teratoma formation in both types of pluripotent stem cells…

2. The Akari’s experimental system is not mimicking clinical use of iPS cells, since mature cells were derived from chimeric animals, but not via cell culture differentiation process. This point was raised by Xu and Knoepfler.

Surprisingly, Akari observed immune response to cardiomyocytes, generated in vitro from iPS cells:

Notably, we observed significant levels of T-cell infiltration when we used in vitro derived cells, cardiomyocytes, as donors (Supplementary Fig. 13).

This sentence, hidden deeply in the discussion section can change conclusions and title of the paper completely!

3. Although sophisticated, chimeric mice as an experimental system could be not the best to address all immunogenicity issues. Embryonic development can induce tolerance or/ and circumvent immunogenicity of donor’s iPS cells:

It is possible, says Xu, that the most immunogenic cells were rejected as the mice developed, which would explain why Abe and his colleagues observed a limited immune response. Transplanting tissues from chimaeric mice is “flawed”, he says.

The same mechanism could lead to silencing of 2 immunogenicety-linked genes, described by Zhao (see comment of Paul Fairchild):

“By generating mice from the iPSC, [Abe’s group] may have unwittingly circumvented this [immune rejection] issue, since silencing of these genes is far more likely to occur correctly during normal mouse development than during differentiation under contrived conditions in vitro,” Fairchild said. And since human iPSCs would have to be converted into other cells in contrived conditions, he added, “I would suggest that the jury is still out as to whether the products of iPSC differentiation will turn out to be immunogenic or not.”

So, both studies have some flaws and we still can’t answer the question of immunogenicity of iPS cells. Finally we have no idea about immunogenicity of human iPS cells and their derivatives.

Please let us know if you know any unpublished reports (from conferences) about reproducibility of Zhao’s findings from 2011.

Tagged as: controversy, immunogenicity, iPS, teratoma