CD34 expression on human mesenchymal stromal cells

by Alexey Bersenev on November 28, 2012 · 5 comments

in mesenchymal

Post to Twitter Send Gmail Post to LinkedIn

The recent review, published in Cytotherapy, challenges the consensus about absence of CD34 expression on mesenchymal stromal cells (MSC). The authors argue that CD34-negative MSC is nothing more than cell culture phenomenon:

The finding of a lack of CD34 expression was thus based on MSC that grow on a plastic surface, not MSC that reside in the bone marrow. Most, if not all, subsequent studies that have identified MSC as lacking CD34 expression were also based on plastic-adherent MSC.

The authors identified human adipose tissue-derived MSC as CD34+. They referred to other studies, which confirm their findings. Indeed, if we look at surface markers profile of adipose-derived MSC, attached to blood vessels, we can detect them as CD34+ in situ (by immunohistochemistry) and freshly isolated in suspension (by flow cytometry). Relative abundance of MSC in adipose tissue allows us to study them in natural environment more precisely. The most sophisticated flow cytometric analysis of freshly isolated adipose-derived MSC was performed by Donnenberg’s group. His very recent study, confirms that freshly isolated MSC of adipose tissue stromal vascular fraction, express CD34.

The ultimate experiment to confirm/ reject the possibility for CD34+ MSC should be the following: sort CD34+ cells from bone marrow/ adipose or other tissue –> culture in MSC favorable conditions –> propagate plastic+ cells –> characterize MSC properties (colony formation, markers exprression, tri-lineage differentiation). Indeed, such experiment were done in 1991, by Simmons & Torok-Storb and in 2007 by Kaiser. They confirmed that pure bone marrow CD34+ cells can give “classical MSC” in culture.

Interestingly, “CD34 story” could be even more complicated than we think. First, there is a difference in CD34 expression on MSC, isolated from different tissues:

Compared with BM-MNCs, all native ASCs were found in the CD34+ cell fraction of the AT-SVF. Native ASCs expressed classical mesenchymal markers described for BM-MSCs. Interestingly, CD34 expression decreased during ASC cell culture and was negatively correlated with cell proliferation rate.

Second, CD34 expression on MSC could switch, based on environmental cues:

Thus the CD34 expression status in HSC and MSC appears to depend on the environment, and can change from positive to negative, and vice versa, as they relocate from one tissue compartment to another. More importantly though, for therapeutic applications, both freshly isolated (CD34+) and cultured (CD34-) ADSC have been shown to be effective in treating various diseases in animal experiments (2,3).

So, the historical published data about bone marrow and recent comprehensive characterization of adipose-derived cells, confirm, that at least some MSC, residing in their natural environment, are CD34+. However, the notion about CD34 positivity contradicts ISCT minimal criteria for MSC, published in 2006:

MSC must express CD105, CD73 and CD90, and lack expression of CD45, CD34, CD14 or CD11b, CD79α or CD19 and HLA-DR surface molecules.

It seem to me that ISCT report fails to mention that these criteria should be strictly applied to cultured plastic+ MSC. Well, maybe it’s time to make a correction:

Thus, based on the evidence presented above, we propose a modification of the minimal criteria for MSC to include a distinction between their tissue-resident and cell- culture states.

I like this review and discussion. The author’s argument looks very compelling to me! I would love to hear your opinions and experience. I know that many readers of this blog have experience with MSC. I’m especially curious to hear about cord blood- and umbilical cord tissue-derived MSC.

{ 5 comments… read them below or add one }

Larry Snyder DVM December 2, 2012 at 9:34 pm

Alexey, As a veterinarian that has done over 70 adipose stem cell cases using MediVet-Americas technology, I am still confused about which of the stem cell markers are the most important for cartilage and bone regeneration. I have always been under the assumption that the CD34+ cells were the workhorses and if that is the case adipose tissue should be superior to bone marrow. We have had such phenominal results on animals with this technology, I am amazed that human usge is lagging so far behind. Thank you for your input.


Alexey Bersenev December 3, 2012 at 3:10 pm

For osteo-, chondro- repair, stem cell markers are not important, but rather osteogeneic-, chondrogenic progenitor cells markers, such us STRO family, Osterix, collagens and so on. Based on general assumption (and at some part consensus) CD34 is a marker of human hematopoietic progenitor/ stem cells and moused hematopoietic progenitors (not stem cells). But CD34 is non-specific for hematopoietic lineages.
There is a general assumption that MSC are CD34-, but it only applies to cultured cells. Now, because we have tools for better cell characterization in adipose tissue and relative abundance of MSC in the fat, studies have been shown that adipose-derived MSC are CD34+ in situ and freshly isolated. It seem like CD34+ MSC in the fat associated with increased proliferation dynamics.
Human studies are lagging behind of veterinary, because of regulatory system. But at some indications human application could be superior. For example, we have 4 commercial stem cell drugs (all MSC-based) on world’s markets.


Belaid Sekkali December 14, 2012 at 11:08 am

Dear Alexi,

Thank you for sharing this very interesting information. Regarding in situ CD34 expression on MSC, what if hypoxic condition (Stem cell niche) is the driving force behind CD34 expression and that under normoxic conditions (cell culture standard) CD34 is lost? Is CD34 an inducible hypoxic gene?



Alexey Bersenev January 1, 2013 at 1:52 pm

I don’t know the answer, but my guess would be yes. It is may be linked to hypoxia, even though > 95% of progenitors, but not stem cells are positive CD34+.
Need to check publications on “CD34 expression and hypoxia”.


Michael Buratovich October 27, 2014 at 8:26 pm

Virtually every paper I have read on MSCs seems to either say straight up or intimate that MSC populations are not homogeneous. Maybe CD34 expression is simply one of the vagaries of a rather variable cell type that has some that are CD34 positive and others that are CD34 negative. The real question, it seems to me, is are there genuine biological differences between CD34 positive and CD34 negative MSCs. That question has not been fully addressed to my thinking. I think it is perfectly reasonable to suspect that the CD34-negative phenotype of MSCs is an artifact of being cultured.


Leave a Comment

Previous post:

Next post: