Searching for a set of molecular characteristics common to all stem cells is, at best, a quixotic quest, argues a systems biologist in an opinion piece recently published in the Journal of Biology. This overly-simplified view of stem cells, the article notes, may be leading science down unfruitful paths and holding back clinical research.

Instead, the author of the review, University of California, Irvine, researcher Arthur Lander advocates a change in mindset. "It is perhaps curious that, after 45 years, we have been unable to place the general notion of 'stemness' on a purely molecular footing," he writes in the article. He posits that the time has come to consider stem cells in broader physiological contexts, studying their role in complex networks of molecular and cellular interactions rather than examining the function and genetic makeup of stem cells in isolation.

Lander took time to chat with The Scientist about reframing stem cell research to emphasize this systems-level perspective and the intricate feedback loops that spur stem cells into action in vivo.

The Scientist: Can you start by enunciating your main point about the fallacy of the stem cell concept?

Arthur Lander: It's a little easy to get carried away and think that I'm trying to say there's no such thing as stem cells, which I'm certainly not. The main point I'm trying to make is that the most prevalent conception of stem cells within the scientific community is of the sort of thing you would expect to have a unique defining molecular signature...[and] we shouldn't necessarily expect that there will be some unique molecular signature of what it means to be a stem cell.

TS: Are you saying that we should abandon the almost 50-year-long effort to define the concept of 'stemness' and change how we think about it?

AL: We want to nail down the concept, but we need to nail it down in a different way. So if this is really a network property -- that is, a property at a higher level than the cell itself -- then nailing down the concept means knowing something about molecular properties of the cell and molecular properties of cell types it produces, the molecular properties of things that they produce that then feed back on the stem cells? It's a system level property, so we need to have information about a whole system.

The [essential] pieces...may be more about the dynamics of responses. For example, you could have one system in which a particular factor is used for the feedback loop and another system in which a different factor is used for the feedback loop. Exact molecular details of what it is -- whether is a TGFbeta family member or notch signaling or a hormone -- those may not at all be defining characteristics of a stem cell system. Those may be peculiarities of each individual case of stem cells.



TS: How do you reconcile the fact that adult cells are capable of being reprogrammed using a very small number of genes? Wouldn't those genes be molecular descriptors of 'stemness?'

AL: I would disagree. Those sorts of experiments are really not addressing the issue of 'stemness.' They're addressing the issue of potency. They're saying you have cells which appear to be unipotent, and you can put in a few genes, and you can make them pluripotent. So I suspect that certain degrees of potency, by themselves, are concepts that can be reduced to particular patterns of gene expression. Whether one needs the same cocktail of genes in every different cell type to get to the same stage of potency remains to be seen. There may be many routes. In a sense, by saying, 'Let's be careful about how we use stem cells,' I'm trying to advocate that people, when they study potency, say that they're studying potency and not 'stemness.'

TS: In your article, you drew the comparison to phlogiston -- the disproved 17th century idea that a common and mysterious substance is released from all burning materials. That surprised me, because phlogiston was something that science found to not exist. What is your comparison between the concept of stem cells or 'stemness' and the mistaken phlogiston theory?

AL: There, the point I'm trying to make is that the concept can have a perfectly good operational definition and still refer to nothing that actually exists. Later on in the article, I bring up the concept of the rate-limiting enzyme as probably the most similar sort of concept to stem cell. So like phlogiston, rate-limiting enzyme is defined by an operation: You look at a pathway, and if you take this thing and change its levels, the levels [of metabolic product] coming out of the pathway change. We can point to all kinds of rate limiting enzymes, and we can purify them, we can characterize them, they have molecular properties.

But what they don't have is a single distinctive set of molecular properties that defines what it means to be a rate limiting enzyme. And that's because the concept of rate limiting enzyme is entirely contextual. Ditto for stem cells: You take them out of their context, which people like to refer to as 'niche' -- and they never behave the way they do in vivo.

TS: You seem to say that the concept of rate-limiting enzymes, as an operational definition, is useful in some ways. Do you feel the same way about stem cell as an operational definition?

AL: Absolutely. Clearly one can define the behavior of being the stem cell in a system. And so operationally, you can talk about, 'Can I transplant some cells that will behave this way?' or 'Within cancer, are there some cells that are behaving in this manner?' The tricky part is if then you jump from that to say, 'Well great, so since there are cells behaving in that manner, I should be able to define a network of gene expression that tells us what it means to behave in that manner.' That would be like saying that I should be able to find a type of fold or a type of amino acid that defines being a rate-limiting enzyme. That, of course, is nonsense.

TS: What is the damage or the danger in perpetuating this mistaken concept of stem cells or 'stemness?' Are there any false paths that we've been led down?

AL: I shudder a little bit about being too critical about false paths, because I think the natural course of science is that we go down lots of false paths. I think we have to. If we try to be so clever that we never went down anything that might be a blind alley, we'd never get any work done.

[Cancer therapy research is] a case that has a certain clinical immediacy to it. You don't want people jumping in and making assumptions about how therapy would work that would lead them to explore some therapies and not others that would perhaps hold back the course of finding better therapies for a certain amount of time. That has real impact on people and their lives. Whether those of us who are doing research on animals or doing theory get held back by a little bit...I'm not sure that that's the end of the world.

TS: Do you think that by shifting the framework away from a single cell to a broader system, there's a danger of the concept of 'stemness' becoming too diffuse? Would we be losing some level of detail by pulling the microscope back from the cell to consider the system?

AL: We might be. I think that just remains to be seen. It remains to be seen how many different ways there are to achieve stem cell behavior within a lineage. If there are 100 different ways and [that number] keeps growing, then there might be a sense in which the concept is a nice concept, but it refers to so many phenomena that it's not that helpful. On the other hand, we might find that there really are a very limited number of tricks that are used, and so the concept might be quite useful, just having been bumped up to a slightly higher level.

TS: Have you presented your ideas to any colleagues or stem cell researchers and gauged their reactions?

AL: I have presented it to some colleagues, and I haven't had anybody yell at me yet. There are others who have been sounding this note for a while now. What's different here is that in the work of others they've tried to argue that stem cell is a condition rather than a character based largely on some of the difficulties that have been obtained in trying to pin down 'stemness.' I come at it by saying that in addition to that, we now know, because of our understanding of feedback control, that we shouldn't expect to be able to pin down stemness. It's just not going to happen unless we treat it at a network level. So I'm trying to add to the chorus here rather than be the lone voice. When I present this, it's not as radical as it might seem, and I hope I don't come across as though I'm trying to take credit for being the only person in this camp.


Related stories:

One step to human pluripotency
[28th August 2009]

Stem cell therapy triggers tumor
[18th February 2009]

Identifying the niche
[June 2007]