Is human VSEL stem cells a real thing?

by Alexey Bersenev on April 21, 2012 · 8 comments

in cord blood, other adult stem cells

Post to Twitter Send Gmail Post to LinkedIn

Very Small Embryonic-Like (VSEL) stem cells is a proposed pluripotent population, residing in adult tissues. It was discovered by Mariusz Ratajczak‘s group in 2005. Since that, VSEL stem cells were isolated from a few mouse and human tissues and tested in disease models. Mariusz Ratajczak is proposing VSEL stem cells as a future of regenerative medicine. Recently, VSEL stem cells platform was commercialized by NeoStem.

However, there is some skepticism among other stem cell scientists about the validity of VSEL stem cells as defined pluripotent population in human. Currently, there are 36 results for a query “VSEL” in PubMed. The vast majority of the original studies is attributed to Ratajczak’s group. In this case, the importance of independent validation of their findings is obvious. So far, I was able to find only one independent publication, which is not associated with Ratajczak’s group. But, this study doesn’t look very compelling to me, because there is no flow cytometry data.

Finally, a new independent study, assessing VSEL stem cells in human cord blood was just published in PLoS ONE. I think, this study is well done and worth our attention.

SCA evaluation:
1. Isolation and study of VSEL completely rely on flow cytometry (FACS). This method is essential and should be presented in details. Unfortunately, FACS plots from the Ratajczak’s published studies are lack of controls (fluorescence minus one – FMO), cell aggregates exclusion, viable DNA+ cells analysis. For cell population, which looks like a debris such detailed analysis is essential. The advantage of Alt’s study is careful assessment of VSEL by FACS. They used FcR-block to exclude non-specific binding, Hoechst 33342 for exclusion of debris from nucleated cells, viability dye (PI), gating for aggregates exclusion, double sorting technique for maximization of purity and minimization of debris and unwanted cell contamination. In comparison with Ratajczak’s studies, populations on FACS plots were more defined. Still, I’d show FMO controls, because gating strategy in this case could change everything.

2. Alt’s group have checked expression of 22 markers on VSEL cells by FACS.

From a total of 22 markers tested, only three were found positive: CD31, CD84 and the previously described VSEL cell marker CXCR4.

They were able to identify the same VSEL cell population, originally described by Kucia in 2007. In contrast to Kucia data, Alt’s study failed to show CD34 and CD133 positivity on VSEL. They showed, that 2 new markers highly expressed on VSEL – CD31 and CD84.
So, even though the existence of VSEL was confirmed, the difference in markers expression could indicate to heterogeneity of the population or difference in cell preparation/ processing.

3. The authors confirmed, that VSEL cells are insensitive to red blood cell lysis and that density gradient centrifugation (by ficoll) does cause considerable loss of this population. Nevertheless, they were able to identify phenotypically the same VSEL population in interphase and pellet after the ficoll procedure.

4. Alt’s group was not able to grow sorted VSEL stem cells in any tested culture conditions. Cells failed to form colonies on OP-9 stroma. Original Kucia’s study didn’t test hematopoietic activity of the described population. But, inability to grow on OP-9 contradicts to very recent Ratajczak’s study. A standard (methylcellulose-based ) colony assay should be done by any group, assessing cord blood-derived stem cells.

5. The transcriptional analysis (microarray) of sorted VSEL reveled the absence of “stemness” markers, including Oct4, Nanog and Sox2. It looks like once cells were “cleaned by improved FACS”, the whole “pluripotency markers” were gone.

6. Unfortunately, Alt’s group didn’t assess “pluripotency markers” by other methods, such as RT-PCR and immunofluorescence (immunohistochemical staining), which were used in Kucia’s study. It will be interesting to compare “pluripotency profile” of VSEL by applying exactly the same methods.

7. Finally, unlike the Ratajczak’s studies, the authors performed karyotype analysis of VSEL stem cells. Surprisingly, they have found that majority of VSEL cells have an aberrant karyotype (aneuploidy and indications to binuclearity, tetraploidy):

The cumulative evidence therefore suggests strongly that hUCB VSEL cells are the products of a defective or uncoordinated cell division, leaving them with extensive karyotypic abnormalities. The associated very low cytoplasmic volume would also be consistent with the observed loss of cellular function.
Even so, the aneuploid products themselves are unlikely to make a long term contribution to regeneration, so that the sum of the evidence presented here argues clearly against a stem cell identity of VSEL cells from hUCB.

This is very interesting findings. I’d agree with author’s conclusions.

Indeed, there has been no independent evidence to date to support that a pure UCB VSEL population possesses functional hallmarks of stem cells, such as in vitro differentiation potential, high proliferative capacity or the ability to self-renew in vivo. It is therefore questionable whether UCB VSEL cells should be termed “stem cells” at all.

8. Overall, this study is good and significant. The authors followed original protocol and confirmed the existence of VSEL population. However the “stemness” was challenged in series of assays. This study is still lacking of functional assessment of described cell population. I think, in vivo transplant model for test of hematopoietic function, should be done.

So, the VSEL stem cell protocols seem to be partially reproducible, but their pluripotency and stem cell function could be under the question. Please, tell us what do you think about this study, VSEL reproducibility and validity. Please, indicate any independent studies, if I’ve missed any.

{ 8 comments… read them below or add one }

Michael An April 22, 2012 at 1:32 am

I don’t have any comment on these cells, but I have a similar question on another cell population, unrestricted somatic stem cells (USSCs) http://www.ncbi.nlm.nih.gov/pubmed/15263023. Are they independently validated and universally accepted?
Thanks,
Michael

Reply

Deepa Bhartiya April 22, 2012 at 10:33 am

I am really surprised that you sideline our study just because we did not use Flow -but that is the beauty of our study. Let anyone use our method –they will get VSELs in the discarded fraction when volume reduction of cord blood/ bone marrow is done. Imagine – cord blood banks are charging so much from new parents to cryopreserve cord blood of baby and unknowingly throwing away the ‘pearls’ during processing. Recent Alt study in PlosOne is a glaring example of what happens when you use wrong cell surface markers and cause confusion … VSELs are lin-/CD45-/CD133+ stem cells.

We first detected VSELs in adult human and mouse testis, then in mouse, sheep, rabbit, monkey and peri-menopausal human ovary surface epithelium (one can follow up our work on Pubmed). After struggling to enrich sheep ovary VSELs for almost a month – we realized that we invariably lost them during processing. These studies/observations led us to quickly examine whether VSELs in cord blood and bone marrow are also lost during processing …. and the doubts were confirmed – which we published. Flow is not essential to isolate the VSELs– if we can avoid flow to isolate VSELs … all the better.

But I am surprised by the topic of this blog – Is human VSEL stem cells a real thing? Of course they are there – they are present in all body tissues… we need to work together and exploit their potential. In fact we have also shown that MSCs and HSCs are actually progenitors ‘descendants’ from VSELs!

Our expertise is embryonic stem cells and we have derived two human ES cell lines. Two major issues of concerns with hES cells are immune rejection and teratoma formation. These are not associated with VSELs – which are pluripotent, can be isolated from autologus source and do not form teratoma.

Thanks
Deepa Bhartiya

Reply

Alexey Bersenev April 22, 2012 at 7:12 pm

Thank you for the comment Deepa!
As I noted in the post, I think, “nice and clean” FACS is essential for VSEL cells identification. On this matter I’d respectfully disagree with you. It was initially describe by Kucia, purely based on FACS and sorting. So, the importance of FACS was postulated in original studies by the group, which discovered VSEL. I believe, that only FACS can isolate VSEL cell at the high purity and therefore exclude background from other assays. This was the main reason, why I analyzed Alt’s study, but not yours. It’s definitely IMHO and I’m very open to any other opinions.
I indicated that absence of immunohistochemistry (IHC) is a weak part of Alt’s study. I’m very curious to compare IHC markers expression from yours, Kucia’s and Alt’s studies.

I don’t understand, why you said “Alt’s group used wrong surface markers”. They used the same markers as Kucia and, indeed, they confirm that VSEL population exists. How would you explain, that the same VSEL cell population in Alt’s and Kucia’s studies have different level of CD34 and CD133 expression? You said your VSEL cells are positive for CD133, what about CD34 and CXCR4?

In regard of “real thing”, I intentionally use provocative title to spark a discussion. As I noted in the post, the existence of VSEL cells seem to be confirmed. What was challenged by Alt’s study is their “stemness” or function. In this regard, as I said, more validating studies should be done, because all of previous (including Alt’s) are lacking functional assays. So, the “real thing” means existence + function. I’d agree with Alt’s group, that all studies are lacking the evidence for self-renewal and pluripotency in functional in vivo assays. For example, Ratajczak’s group said that CB-derived VSEL cells posses HSC activity, but they never (to my knowledge) test it in serial transplantation assay. In terms of pluripotency, markers expression alone could not provide enough evidence. If VSEL are not teratoma-forming, they should repopulate and differentiate in all tissues after injection into blastocyte. Or at least, analogy of Pluritest should be done. It’s striking to me to see that Alt’s VSEL are so much different from ESC and iPS on microarray, despite the similarity (to Kucia’s study) on FACS.

Do you have published data on MSC and HSC as descendents of VSEL cells? What assays did you do to confirm that?

I was happy to see your publication in open access. Also many of Ratajczak’s articles in open access. It was my hope that many research groups would like to try and validate original studies. I guess, we can assume that your studies are confirming Ratajczak’s conclusions. The Alt’s study is challenging “stemness”. I wonder if any other groups done VSEL cells protocols.

Finally, I can provide a platform (this blog) and would like to invite you to share a detailed protocol (in Nature Protocols style) for VSEL cells isolation and IHC.

Reply

Deepa Bhartiya April 23, 2012 at 8:55 pm

Please refer to the latest paper by Ratajczak’s group (www224984.ncbi.nlm.nih.gov/pubmed/52) to define human CB VSELs. Thus we feel that the cells Alt’s group has studied are not the correct starting material.

If we compare pluripotent markers in Layers 2 & 4 (refer to Fig 1 in our paper) – they are all expressed several fold high in layer 4. OCT-4A transcript (responsible for pluripotency) is expressed only in Layer 4. Mouse VSELs are very well worked out – data on human VSELs is emerging.

Pl also refer to the site http://www.ncbi.nlm.nih.gov/pubmed/18536651. This is another group which has isolated VSELs from human cord blood and differentiated into neurons and earlier into hepatocytes.

We have shown in our paper that the VSELs in Layer 4 have nuclear OCT-4 whereas few cells in Layer 2 always showed cytoplasmic OCT-4 – using a polyclonal OCT-4 antibody. It is well established fact that OCT-4 is a pluripotent transcription factor present in the nucleus – and when the cell starts differentiating – OCT-4 is no longer required – shifts to cytoplasm and is eventually lost as cell becomes more differentiated. Similar staining pattern was observed in umbilical cord tissue MSCs (Fig 1 D & E in our paper). We have recently confirmed our findings in cultured MSCs from mouse bone marrow.

Thus ours is a very simple approach – which can be easily adapted in any cryobank – rather than use a sophisticated instrument – …. MSCs have always been contaminated with VSELs – which explains their ‘pluripotent characteristics’.

Another observation of Alt group was that the cells stained very strongly with Hoechst 33342 – but if they were truly pluripotent – they should have excluded Hoechst. We have observed that the VSELs do not stain very well with DAPI – and we have argued that this is because they have abundant euchromatin … whereas DAPI preferentially stains heterochromatin. Also we have observed that progenitors stain very dark with DAPI .. thus possibly Alt group have picked up a population of progenitors rather than the pluripotent VSELs themselves. But whatever the cell type – their results of aneuploidy detection by FISH are indeed surprising

Thanks

Deepa

Reply

Alexey Bersenev April 23, 2012 at 11:49 pm

Deepa,
Thank you for response!

If you think that “Alt’s group was working with wrong starting material/ population”, you’re making a good argument.
But how do you think Lin-/CD45-/CXCR4+ Alt’s population from ficoll pellet could be different from Kucia’s VSEL population? To me it looks the same. Both did a ficoll, lyse RBC in pellet and used the same markers on FACS.

Can you please fix the first link you gave? Is this one?
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3135663/?tool=pubmed

Despite all of markers story, what have been done in the field to prove functionality of VSEL stem cells? Any self-renewal assays? Any pruripotency tests in vivo? Regenerative and therapeutic potential is not necessarily linked to “stemness”, but frequently delivered via progenitors function.

In terms of Hoechst 33342 for nucleated cell versus side population staining, I have to check. Probably different concentrations or protocols used for these assays.

Reply

Deepak Modi May 25, 2012 at 11:37 am

Dear Alexey and Deepa
I have not read Alts paper completely and I do not want to get in to the nity-grity of protocols and concentrations and all. I was wondering that why is it so necessary to show that VSELs HAVE to express ALL the so called pluripotency markers. The VSELs by the nature are Different cell types differing in morphology, size, site of availability as compared to ES cells and hence I am not surprised that they do not express ALL the known markers of stem cells. This is evident from the Alts array data where they show that VSELS have a profile different from all the cell types studied. Thus this is a unique population of cells that needs to be studied.
Similarly for expansion. ES cells could not be expanded under several conditions for a long time till Evans discovered the need of LIF and feeder. As of now we are unable to proliferate does not mean they CAN NOT indefinitely proliferate.
A quick glance at Fig 6 of Alts data makes me think that the cells isolated by them may not be VSELs. The sizes are so close to the other cells whereas VSELs are suppose to be around 2-4 microns as opposed to other blood cells that are always more than 10microsn. The sizes for VSELs and other cells look almost identical.
I was having a discussion with Deepa yesterday and have a gut feeling that the cord blood cells that are so called VSELs are not the true cells but a differentiated form of original ones. IF we assume that the VSELs are quiescent type of pluripotent cells that differentiate at sites of tissue injury, their properties WILL change when they start to proliferate. In the cord the numbers of VSELs are almost 10 folds higher than blood and this could occur only if they proliferate. IN the event these cells are programmed to be quiescent in their native state, trying to proliferate them (in vitro) or picking them up at the state they are proliferating (like in cord) would definitely mean they will change their phenotype and yield conflicting data.
The data shown by Alt on aneuploidy is shocking. I will take Hoechest data with a pinch of salt as at stages when the chromatin is in different states of folding the dye staining may become variable. If we consider the dye data with FISH it appears that these cells (?) perhaps are only thrown off DNA with some cell membrane like the micronuclei being thrown off. However alhough on face value the FISH data looks shocking, Alt has not given any statistics of the numbers of cells they found aneuploid in the whole population. For those of us who routinely do FISH we know that a subpopulation of cells do show aneuploid due to hybridization failure, overlapping and split signals (often happening in dividing cells). Without a clean data on the hybridization efficiency and background data on false positives and false negatives we must take this data with a pinch of salt.
I think we need to have an open mind and carefully note the characteristics of these small cells point by point over time rather than jumping to major conclusions. Clearly we need more data and more uniform protocols and more characterization to reach some conclusion. Till then trying to prove any party right or wrong would be being biased. .
Finally, I really appreciate your effort in starting this blog and providing a platform for the researchers to share their opinion and giving an opportunity to passive followers like me to have an open perspective of this new cells in our body !!

Reply

Jeanne Adiwinata Pawitan May 18, 2014 at 4:31 am

Dear Deepak and Deepa,
I do not do research in VSEL, but I am very interested in VSEL cells, and I have some questions:
1. For Deepak: you said that the size is 2-4 microns, so is it smaller than erythrocytes, which is around 7 microns?
2. For Deepa: a. You said that VSEL reside in the discarded portion, do you mean the upper layer on the top of buffy coat after ficoll centrifugation?
b. In your opinion, will this also apply in bone marrow aspirate?
c. will you please send me your publication concerning the isolation of VSEL from the discarded portion? my email is: jeanneadiwip@fk.ui.ac.id or jeanneadiwip@gmail.com

Thanks a lot for the information

Reply

Deepa Bhartiya May 19, 2014 at 12:14 am

Yes- VSELs are smaller than RBCs. They settle along with the RBCs during Ficoll-Hypaque centrifugation step – it applies to both cord blood as well as the bone marrow.
http://www.ncbi.nlm.nih.gov/pubmed/21780911

Reply

Leave a Comment

Previous post:

Next post: