Functional hematopoietic stem cells from human ES cells – significant progress toward?

by Alexey Bersenev on July 17, 2008 · 8 comments

in Journal club, niche, under discussion

This title is a citation from the last paper about making functional hematopoiesis (hematopoietic stem cells) from human embryonic stem (ES) cells, which appeared in July’s issue of Cell Stem Cell journal. I put ? mark at the end for a reason that i’ll discuss here.

Paper is very good and compared to others from the same topic, definitely advanced. But when i read a comment I asked myself some global questions, such us: Do we really need to produce hematopoietic stem cells (HSC) from ES for clinical needs? And what is the progress in this area in the last decade? I’d like to discuss here about it with people who are really interested or working in this field.

Let’s start from an overview of the problem. The first question: why do we need to make HSCs from ES cells? The answer is that it could be a therapeutic alternative to bone marrow transplant (BMT) in hematological clinic. To become an alternative method, it should have definite advantages compared with others. So, yes we really need more HSC in BMT clinic, because it’s frequently not enough in many available grafts (such as bone marrow or cord blood samples or mobilized apheresis products) for successful engraftment and to rescue patients. So the first point is the number of functional HSC in one transplant – we need expansion! The second point is immunological compatibility – transplant should be a match! Third point is availability of the transplant.

Based on these advantages many scientific groups around the world are trying very hard to improve protocols to make it work. The main problem was to get good engraftment of HSC derived from human ES cells. Seems like cells were not functional in the first attempts to transplant into immunocompromised mice. Only recently in 2004-2006 Mickie Bhatia and Dan Kaufman groups got engraftment of human ES-derived HSC close to 1% only after intra-bone BMT.

Ok, now let’s come back to the most recent work of Maria Ledran,
published in Cell Stem Cell. This group has significantly improved the protocol of HSC derivation. The most important achievements of this work are the following:
* authors successfully mimic HSC niche conditions in vitro, using stromal cells. So it seems like for normal function HSC required natural microenvironment;
* study highlighted role of transforming growths factor beta in efficient generation of functional HSC from ES cells even without stromal cells; thus shows that not only adhesion signaling is important but also humoral factors, released by stromal cells withing the niche;
* improvement of cell culture conditions allowed engraftment efficiency of 2-16% (that’s much higher than was reported before); HSC were functional – and gave long-term multilineage reconstitution and were able to self-renew.

Colony assay, demonstrates hematopoietic progenitor capacity of human ES cells in vitro:

(Credit: Maria H. Ledran, et al. 2008; modified)

Ok, now we have functional HSC from human ES cells, without genetic manipulations with up to 16% of engraftment level (we jumped from 1% hooraay!).
I’d like to notice that this study was done on NOG mice (immunocompomised mice with very low innate immunity and absence of NK cells) and unfortunately didn’t compare efficiency of human somatic HSC engraftment in the same model. But if we look at previous study of Bhatia group it shows that engraftment of ES-derived HSC was worse than that somatic HSC from cord blood, and even more – they look different on molecular level.

To give a green light to derivation of HSC from ES for clinical needs we have to compare all advantages and disadvantages of this method with the qualities of current techniques. So authors pointed out that potentially new source of HSC could has enough number of functional stem cells, less immunogenic or even autologous and easily available. I’d challenge all of these advantages and indicate what kind of bad things can we expect from bringing this technology in clinical future:

* we can get a lot of HSC, but longer cell culture time and crazy expansion (compared with expansion of somatic HSC) level will increase risk of mutation acquisition and leukemia development after transplant;
* HSC derived from allogeneic human ES will be immunogenic (maybe a little less) after transplant anyway;
* we can create autologous cancer-free HSC product from ES cells if we will be able to make somatic cell nuclear transfer or iPS techniques work for patients, in this case I don’t think high price of procedure allows for it to move to wide clinical use;
* we should enrich or even sort out HSC contaminated by undifferentiated ES cells or other unwanted derivates (such as endothelial cells which could be immunogenic) to make sure there will be no risk of tumors growth after transplant, which will make procedure more complicated and costly;
* human ES-derived HSC are functionally different from somatic HSC by unknown reasons and have more potential risk of complications after transplantation ( lung capillary emboli for instance).

About a hundred protocols were published about expansion of somatic (adult) HSC (cord or mobilized blood) for clinical needs. Some of them are currently on clinical trials and really work. To overcome compatibility problem, on the one hand we have banks of matched grafts and on the other hand we know or can develop some technologies of separating malignant cells in bone marrow from healthy HSC, so adult transplant could be autologous.

Overall i can not get the points of huge money and time investment in derivation of HSC from ES cells for clinic. Expansion of somatic HSC almost there and I honestly think that we will get only more problems with ES-derivates (even autologous). Of course i don’t think it’s a waste of time, because it gives us a lot of scientific information about hematopoiesis and development. It’s very applicable in basic science or maybe for drug screening. But do we really need it in terms of clinical perspectives? What do you think?


read more about this work:
Growing blood in a dish; using stem cells to screen for breast cancer (by the Niche)
The Origins of Blood: Induction of Hematopoietic Stem Cells from Different Sources (Cell Stem Cell comment)

{ 7 comments… read them below or add one }

Alex July 18, 2008 at 3:27 am

one more study (just came up from Dan Kaufman group), highlighting difference between human ES-derived HSC and somatic (adult) HSC from cord blood:

Here, we demonstrate that despite expressing functional Notch-1, CD34+ cells from hESCs did not derive T cells when co-cultured with OP9 cells expressing Delta-like 1, or in fetal thymus organ culture. hESC-derived CD34+ cells also did not produce B cells in vitro. In contrast, CD34+ cells isolated from UCB routinely generated T and B cells when cultured in the same conditions.
Together, these results demonstrate fundamental differences between hESC-derived hematopoietic progenitors and analogous primary human cells.


Alex July 22, 2008 at 5:19 am

BBC release:
New hope for bone marrow patients

Edinburgh experts used blood stem cells from mice to mimic how humans produce the stem cells and found they were able to multiply them by 150 times.

Would you inject yourself cells that you expanded in petridish for 150 times?? I won’t!

We need expand cord blood HSC only 2-3 times from one sample to make patients happy and it’s already work in clinic!

I’d agree with Dr. Medvinsky – it’s very long way to go. I’d add: so long, better concentrate to another – real things…


Li Xiao August 14, 2008 at 6:55 am

Great article! Thanks!


Todd January 8, 2009 at 4:28 pm

I really can’t believe that such an uniformed article was written by a physician-scientist. I’m not even sure where to begin with all of the false logic and misinformation, but I’d like to point out just a few key points.

First, the differences seen between somatic and ESC-derived HSC are interesting on many levels, and provide mechanistic development models for the blood cell lineage specification. With the recent developments in iPS cell generation, ES-HSC issues become even more important as scientists apply lineage specification to iPS cells from patients with diseases in order to study the disease etiology. Discoveries made while examining the inefficient ESC-HSC transition will be applicable to iPS cell differentiation as well as onset of malignancy. I’d say those are worthwhile reason for “why do we need to make HSCs from ES cells?”

As to your second point, you should have read more than one paper from Bhatia, or Zavazava in PlosOne, or countless other papers detailing induction of immune tolerance by cells derived from ESC, a result confirmed in mouse, rat, sheep, and human Bhatia specifically details the immune-privilege of ESC-derived HSC. Further, application of the technology to iPS cells would result in perfectly donor-matched transplant cells, thus obviating concern for immune tolerance.

To false logic, you maintain that cell division=mutation=cancer, yet your own body is replenishing your gut, skin, and yes, even your very own blood cells from a pool of stem cells in each system that expand thousands of fold during your lifetime. Wow, that mitosis sure does work well.

Finally, to plain old misinformation, CD34 expansion in vitro is NOT robust, and is NOT the clinical standard for transplantation; in EVERY case, donors will be mobilized multiple times versus ex vivo expansion. Additionally, cord blood does not yield adequate CD34 cells for transplantation of anyone past a few months old. The only successful clinically-relevant results to date have been with work using mixed-cord transplants, which face the same immunological difficulties that you don’t understand very well. And finally, I’m sure researchers will be thrilled to hear that you ‘know or can develop some technologies of separating malignant cells in bone marrow from healthy HSC, so adult transplant could be autologous.’

I’ll notify the Nobel committee and start enrolling patients in your trial.


Alex January 9, 2009 at 4:07 am

Before i posted this article I had collected different opinions about this problem based on publications and lectures. So there is a controversy going on here. Because I’m not doing this particular type of research and I’m not a practicing physician right now, i was curious what if i’ll write some introduction, my thoughts and doubts and ask specialists to comment and express their opinions here in order to find some truth.

I’ve read more than 1-2 article (I cited here only 3) and I am familiar with Dr. Zavazava’s (PlosONE) paper and induction of immune tolerance by ES-derivates. I don’t think used any false logic. When I wrote about possible mutations I wrote “increased risk of mutation with increased expansion”.

Also I didn’t make any claims that expansion of CD34+ cells is robust and it’s a clinical standard. I pointed out instead that it’s on the way to clinic because there are many clinical trials going on last 5 years and right now and next 2-3 years more likely we will get an approved clinical-grade expansion protocol for BMT patients.

Double unit cord blood transplantation works very well in clinic right now, but nobody has gotten close to the mechanism of how samples compete for engraftment yet – that’s true. Also, except expansion, we have mobilization technique and intra-bone BMT. More of that, pharmacological methods of increasing engraftment on the way.

About “Nobel comittee” – everything that I said in the post is based on published or reported findings and I can link to them. In particular about this technology you can see the lecture of Irv Weissman (min. 11-13). So notify the Nobel comittee about him, I think Weissman is worth it.

So, overall I still think that improving technologies for somatic HSC transplantation have a better chance of success than hES-derivates in clinical settings.

Also I have to notice, that after I posted this article, I didn’t get any feedback and replies that I wanted, there was NO discussion here. I guess that physicians and scientists mostly don’t care (or are too busy to care) about whatever blog posts and maybe nobody is taking it very seriously. That’s why for me being a little bit provocative is one of the ways to facilitate discussion. I even linked my post to some prominent scientists in the field via email and i got some replies, but they didn’t go to the blog.

I hope you can tell us about your findings or give us some links which confute my thoughts and we can get close to the point of discussion.

Finally, I want to note that you can be as critical as you would like to, but be respectful to the author in your comment.


pam84 July 23, 2010 at 3:41 am

I think it’s convenient to develop these techniques. I agree that HSCs derived from allogenic ESCs wont be inherently superior to allogenic HSCs from other sources. But presumably they’d be easily expandable. which could prove an advantage, particularily if safe , stable ESC lines are developed and used for this goal. I recall that ISSC had this idea (or at least commented often in their press releases), to use it’s parthenogenesis-generated SCs to keep huge cell banks, and use them to generate allogenic but hopefully HLA matched cells for patients. Presumably a simmilar idea could be applied in this case, and would more or less guarantee the avaiability of closely-matched transplants.

Also, it’s worth noting that, whereas ips and SCNT are not yet ready for the widespread clinical use, this might not always stay so, and hence one day it might well provide an option to perform autologous HSC transplants in patients in which it would have been very difficult or impossible otherwise.


pam84 July 23, 2010 at 3:54 am

One final note: Leaving aside whether these technologies would become mainstream or not, it’s still convenient to have alternative cell sources for treatment and therapy.


Cancel reply

Leave a Comment

Previous post:

Next post: