Stem cell journal club: dishing on Nature paper on making iPS cells inside mice

in vivo iPS cellsWhat if you could reprogram cells inside of an organism to a different fate?

We can, right?

But when most of us think about making induced pluripotent stem (iPS) cells, we imagine it all happening in a little plastic dish in our labs or in our colleague’s labs, not in an actual organism.

A cancer research team at the Spanish National Cancer Research Center took a very different approach reported in a hefty Nature article. They made iPS cells inside of a living mouse.

They call these “in vivo iPS cells”.

A novel, cool aspect to the article is that the authors present evidence that the in vivo iPS cells are not just pluripotent, but also possess totipotency where they could make extraembryonic tissues too. That is a fascinating development.

The piece, Abad, et al., is definitely intriguing, but there are some significant caveats as well, particularly as it relates (or not) to clinical relevance for regenerative medicine.

The group, led by PI Manuel Serrano, used a dox-inducible system in mice (aka “reprogrammable mice”) to turn on expression of reprogramming factors. They used the classical Yamanaka 4-factor set: Oct4, Sox2, Klf4, and c-Myc.

Switching the 4 factors on had dramatic effects in mice. Too much of the factors (either in quantity or time) wiped the mice out. They all died. There were massive numbers of teratoma tumors.

The researchers found somewhat of a sweet spot where the mice survived to the extent that they could be studied.

Most of the tumors inside the mice appear to be bona fide teratomas indicating that somatic cells in the mice were reprogrammed to a pluripotent state.

Interestingly, the spatial presence of teratoma was not random but was highest in abdominal organs and fatty tissue (see Fig. 1g above). This suggests differential sensitivity to reprogramming. That’s thought provoking. I wonder why some organs did not reprogram significantly? It is also notable that some organs only had mature teratoma.

It’s also worth noting that many of the teratomas were immature, a type that is far more dangerous. This also raises one of caveat about the paper: malignancy.

While they report on some non-teratoma tumors in Extended Data Figure 4, the authors make no mention of malignant teratocarcinoma so I wonder how they could be sure that the immature teratoma-like tumors they observed were indeed “teratoma” with immature components versus the far more ominous teratocarcinoma? I see no data to resolve this issue and would have liked to know more.

If indeed reprogramming produces, as one would expect, significant numbers of both malignant and benign tumors, then that reduces the clinical relevance of this paper in humans. Massive numbers of benign teratomas are clearly fatal in it of themselves and if you throw in some malignant cells in the picture too, then you have a pretty scary clinical picture.

A Nature News piece on this development is subtitled “Reprogrammed cells open door to regenerating tissues in their natural environment,” but does this paper really lay the foundation for that kind of future approach? I’m not so convinced. The paper did not document regeneration, only undesired teratomas.

Another caveat is that of course in vivo reprogramming has been reported in the past so that reduces the novelty of this paper somewhat.

The previous in vivo reprogramming work that I’m referring to is the study from Doug Melton’s lab from 5 years ago also in Nature entitled “In vivo reprogramming of adult pancreatic exocrine cells to beta-cells”.  In fact, there are quite a few other papers with “in vivo reprogramming” in their titles on Pubmed and only some are reviews. Other groups have achieved in vivo reprogramming and reported it too.

The consensus is that the most promising in vivo reprogramming is not all the way back to a pluripotent state, but rather is of the kind that Melton did to make beta-cells. This “direct reprogramming” approach (what we used to call “transdifferentiation”) appears far more practical and logical than making in vivo iPS cells.

You can expect to see more papers like this with different in vivo reprogramming regimens such as leaving out Myc. It will be fascinating to see what these future studies report.

Overall I found this paper quite notable on a number of levels, but like all papers it has some issues as discussed above.

What’d you think? What did you like and what concerned you about the paper?

13 thoughts on “Stem cell journal club: dishing on Nature paper on making iPS cells inside mice

  1. Pingback: Yamanaka in vivo | Stem Cell Assays

  2. Pingback: Knoepfler Lab Stem Cell Blog – Stem cell journal club: dishing on Nature paper on making iPS cells inside mice | Stu's Stem Cell Blog


  3. This is the type of information I like to read on your blog. I hope you will continue to update us on the research side of the industry which is your forte. Much better than speculating on stem cell clinics and what patients should and should not do when it comes to their own healthcare decisions. I truly enjoyed this article. Thank you.


  4. I agree VERY interesting read! Thanks for keeping us updated on iPS cells. “direct reprogramming” approach, what we used to call “transdifferentiation, appears far more practical and logical than making in vivo iPS cells” Fascinating!


  5. Hi Paul: My lab group chatted about this paper over lunch. This has been a lunchtime-and-hallway subject in my lab for a couple of years- with the extremists wondering whether it would make the mouse sort of melt into one big tumor. But we never could think of a reason for doing it other than for satisfying our curiosity. So, this paper does satisfy our curiosity, so we won’t keep talking about doing it. But no one doubted that there would be lots of tumors-so I’m unclear about how the results will have impact on anything we do.. I don’t see any insights that inform us about clinical use of stem cells.


    • Thanks for the comment, Jeanne.
      I very much felt the same way. The paper is one of those that reports on a project that is kind of nifty, but I don’t find the results overall particularly novel or instructive, esp. from a clinical perspective.

      Do you believe that the in vivo iPS cells are totipotent? Perhaps that was one surprise?


      • Hi Paul:
        Human iPSCs and ESCs can certainly make extraembryonic cell types…I guess I’m just not excited about mouse iPSCs being able to do that. People have made extraembryonic cell types from mouse ES cells. Just recently: Nat Protoc. 2013 Jun;8(6):1028-41. doi: 10.1038/nprot.2013.049. Epub 2013 May 2.
        Derivation of extraembryonic endoderm stem (XEN) cells from mouse embryos and embryonic stem cells.


  6. This experiment is insane – they used uncontrolled expression of “Yamanaka 4 factors” yielding wide spread teratoma production – this is not in-vivo reprogramming per the title of the paper – it’s in-vivo “bolus de-differentiation” – reprogramming requires combining this process with a bona-fide re-differentiation process – this is precisely why salamanders and other organisms can regenerate flawlessly and we don’t – another example of the reductionist model looking at the problem with too fine a detail without understanding at the big picture (teaching birds how to fly)


  7. Hi Paul,

    Don’t you find peculiar that the in vivo teratoma locations are the sites that are more susceptible to develop aggressive fatal cancers (Pancreas, Intestine, Stomach…)?? is that just a trivial random phenomenon or something that could be more specific?

    If possible, would like to hear your comments or thoughts from the discussion feed.

    thanks again for making an open “journal club” about this fascinating paper.

    All the best

    Spock


  8. Paul,

    I’m skeptical about the supposed representation of result of “in situ” reprogramming from the cellular microenvironment via the culturing of blood bourne iPSCs. If reprogramming is truly in situ, would there not be inherent differences in the cellular microenvironments (or the various organs) that may even be responsible for the differential expression of teratomas?

    However, I am surprised at the contrast in terms of gene profile between in vivo iPSCs vs. in vitro iPSCs and ES cells. Though I agree this carries little clinical relevance (asides from total avoidance), as most transplantation occur post-reprogramming.

    Additionally, I wanted to point out another paper that was just published today, I was hoping you could give your insight on this paper.

    Y. Rais et al., “Deterministic direct reprogramming of somatic cells to pluripotency,” Nature, doi: 10.1038/nature12587, 2013.

    Thank you,
    James Hong

Comments are closed.