Induction of mesenchymal-to-hematopoietic transition in human stromal bone marrow cells and fibroblasts

by Alexey Bersenev on June 26, 2011 · 2 comments

in hematopoietic

Post to Twitter Send Gmail Post to LinkedIn

The recent study, published in PLoS ONE, describes a new methodology for derivation of blood progenitors from stromal bone marrow cells (MSC) and skin fibroblasts. The authors used chemical 5-azacytidine (Aza) and growth factors to induce mesenchymal-to-hematopoietic transition:

We obtained both mature and immature hematopoietic cells from transformed mesenchymal and skin cells. Transdifferetiation of MSCs and skin fibroblasts generated a mixed population of mature and immature cells such as megakaryocytes, myeloid progenitor cells and stem/progenitor cells capable of engraftment in a xenograft mouse model, suggesting that the transdifferentiated cells might also be suitable for HSCT.

Interestingly, the mechanism of induced mesenchymal-to-hematopoietic transition is activation of HOXB genes family:

Reproduction of similar results with normal bone marrow MSCs confirmed that this procedure is not cell line restricted and might be clinically applicable. It also raised the possibility that the same molecular mechanism(s) might be recruited to induce mesenchymal-to-hematopoietic transition in both HS-5 and normal marrow MSCs. To elucidate the Aza plus GF transdifferentiation-inducing molecular mechanism, both mRNA and DNA methylation array studies were conducted, and analysis of the data suggested that HOXB transcription factors might be operative in this process

HOXB4 overexpression can induce hematopoietic stem cell expansion, but in the same time trigger and facilitate leukemogenesis.

I think, this approach is quite interesting and, potentially, could be a good alternative to direct transdifferentiation induced by defined genetic factors. But some conclusions from this study could be premature.

SCA evaluation:

  1. The authors didn’t assess hematopoietic stem cell function properly. Time point in 3 weeks is not long enough to conclude about long-term repopulation ability of human HSC in xeno- model. The same problem with serial transplantation assay.
  2. The authors should use combination of greater number of surface markers in order to conclude the possible presence of human HSC after engraftment in mouse bone marrow.
  3. Primary MSC and fibroblasts, transdifferentiated by Aza and growth factors, were not assessed for repopulation activity via xenotransplant.
  4. Hematopoietic cells, expressing bunch of different markers, sometimes could not be indicative for their function. Surface markers expression, induced artificially, could be just an artifact. So, if the authors said that they derived megakaryocytes of lymphpoid cells, they should assess them functionally.
  5. The authors indicate downregulation of CD90 as a marker of MSC with chemical induction. But this marker is also known as positive for hematopoietic stem cells discrimination.
  6. The only conclusion we can make is that hematopoietic progenitor cells were generated efficiently. This conclusion based on markers expression (CD45+/CD34+), colony assay, and transplantation assay (repopulation of conditioned recipient).
  7. More sensitive xenotransplantation model (SNG or NOG mice instead of NOD/SCID) could be used in order to test efficiency of the method.
  8. Finally, the possibility of potential leukemogenesis by increased doses of “induction factors” should be assess more carefully by prolonged observation and serial transplantation.

What do you think about this study?

{ 2 comments… read them below or add one }

Mohan C. Vemuri June 27, 2011 at 6:16 pm

This study is a good academic exercise to understand Mesenchymal-to-hematopoietc transition of cells. As has been clearly summarized (1-8), this study and its ramifications have skewed liabilty than reward, but also indicate the sensitive nature of tumorigenesis, due to the upregulation of HOXB4 and other Hox family gene members. Potentially, the well studied mechanisms of HOXB4 gene regulation by Sauvageau G in original studies do not favor the use of HOXB4 mediation, despite acceptable engraftment , as tumor formation was noticed in non-hematopoietic secondary transplants.


Gary June 28, 2011 at 1:58 pm

You could put 5-azacytidine (or any demethylation chemical) on your breakfast cereal and get induction of all kinds of genes because the entire genome is opened up for transcription. The resulting cells are in no way transdifferentiated. They need to stop publishing this alchemy nonsense.


Leave a Comment

Previous post:

Next post: