"Stem" Cell Types (Embryonic vs. Somatic)

I have been aware of the undifferentiated cells in the embryo - and I assume that these are what are now being call “embryonic stem cells”. Is this correct?
Then somebody was poking around and found what are now also considered “stem” cells - and to keep track, the newly-discovered are “somatic stem cells”.
After typing all this it occurs: Are somatic = somatic = somatic, or are there differences among them. If so, what kinds of differences?

Given the huge noise by morons in this country, there are damned few embryonic cells, so we are now mostly using somatic cells. Correct?


  1. Can either of these be cultured indefinitely? If we get one good culture of embryonic cells, can an infinite number of cells be produced - at least in theory? How about in practice? Somatic cultures?

  2. Are these two types of cells actually identical in function and application? It seems obvious to me that an embryonic cell could be convinced to become any type of adult cell - are the somatic cells predisposed to become whatever tissue originally surrounded them, or does it look like, should someone figure out how to manipulate one “stem cell”, the same process can/should be able to work the same way in both types?

While someone thinks of a better answer to your exact questions, give this a read: http://stemcells.nih.gov/info/basics/pages/basics4.aspx

Thanks - trying to hack through Wiki was getting nowhere.

I don’t mind having to google every 20th word or so, but every word longet than 6 characters is a bit overwhelming.

I wrote up a long response that got eaten, and I don’t have the energy to retype, so here’s a shorter version:

Some background:

Somatic stem cells were discovered first. There are many, MANY types of somatic stem cells; most exist only in the embryo or for a short period in infancy and make a few cell types of a specific organ. A small number of types of somatic stem cells exist in the adult and help with repair (like wound healing) or tissue turnover (in growing tissues like skin, intestine and hair). The first stem cell type discovered in mammals was blood stem cells (“hematopoetic stem cells”), a somatic stem cell type which lives mostly in the bone marrow and makes most types of blood cells. (A type of HSCs also live in the umbilical cord, which is why people now bank their newborns’ cord blood.)

Embryonic stem cells (ESCs) were discovered in mouse research in the 1980’s. They are a type of stem cell that doesn’t exist in nature; basically, if you take the part of the very early embryo that will go on to make the embryo itself (the epiblast), and culture it in the right conditions, you can turn the cells into ESCs. The exciting part about ESCs is that they can make any cell type in the body, unlike somatic stem cells which are more specialized and can only make a limited number of cell types (in some cases only one). Note that although embryos contain many types of stem cells, only this one semi-artificial cell type is called the “embryonic stem cell”; the others are just stem cells that are embryonic. :smiley:

Embryonic stem cells can in theory be cultured indefinitely (see below), but making them in the first place requires destroying an embryo; so, due to the pro-life crowd, in the US federally-funded research on human ESCs is limited to a short list of lines (mostly those that existed prior to the regulations being enacted). Of course, for other species, like mice, you can make as many new ESC lines as you want whenever you want (assuming you have the time and skills).

However, the new hotness is “induced pluripotent stem cells” (iPSCs), which are ESC-like cells that also only exist in culture, but can be made (via special voodoo) from regular boring non-stem-cell cells, like those obtained from a skin scrape or blood sample. One big advantage of this technique (besides no embryos) is that in theory you could take a skin scrape off a patient and make them a personal iPSC line from their own cells, which could then be used to make cells for transplant that wouldn’t have rejection issues. A lot of ESC research has shifted over to iPSCS, but there are still various complicated reasons why ESCs might be better, so they haven’t been completely abandoned.

So, your questions:

ESCs and iPSCs can in theory be cultured indefinitely, but over time they will accumulate mutations; if you are careful you can minimize this, but in practice you probably can’t completely avoid it. Some somatic stem cell types can be cultured indefinitely, but most can be cultured for only a short while or not at all. In practice, since humans vary pretty widely and every single one of us carries a wide variety of potentially harmful mutations and genetic abnormalities, if you want the cells for either medical treatment or just research it helps a lot to have multiple options, so even if you have one great line that grows like gangbusters you still need other lines. Again, this is a strength of iPSCs; you can make them from anybody, anytime (given a couple of months, a skilled researcher and some pretty involved techniques).

As mentioned, there are a large number of somatic stem cell types, they vary widely, and most make a limited number of cell types. There is research aiming to take an easily-accessible somatic stem cell type, like skin stem cells, and persuade them to make cells they normally wouldn’t; there is also research aiming to take a non-stem-cell type and covert it directly to a different cell type without going through a stem cell stage (“direct transdifferentiation”). So far there is some success with both of these, but these techniques are less generalizable than iPSC / ECS work (figuring out how to get skin cells to make pancreas cells won’t help you much with figuring out how to get skin cells to make nerve cells).

iPSCs are pretty similar to ESCs, so protocols that have been established in ESCs have usually worked in iPSCs.

A small point that I’m not sure is due to confusion or just a typo: a “somatic cell” just means a cell that isn’t part of the germline (which includes the cells that make sperm and eggs, and the stem cells and precursors that make those cells). Most somatic cells are not stem cells; if you mean “somatic stem cells” you need all three words.

Thank you for such a detailed and thoughtful response!

A quick google shows people stating that there are at least 70 existing therapies using cord blood (a random sample

This sounds a whole lot like a great deal of time, money and (exceedingly hard to find) expertise is being expended on adult somatic stem cells because ESCs are essentially illegal to produce due to religious objections* Fair conclusion?

Is the purpose of the manipulation of adult cells to get them back to the same state that the pre-differentiation embryonic cells?

How common are stem cell therapies in the field? I have both kidney failure and osteoarthritis and I’ve never heard even the vaguest of hints that there might, someday, somewhere be a treatment.

    • it is one thing to assert that an unplanted egg in a tube becomes a full-fledged human the instant a sperm penetrates it. It is stretching things just a bit to argue an egg in a petri dish is a person. Sigh… I do hope there are healthy young researchers with unlimited supplies of petri dishes…

p.s. - I keep a blank document up when typing long posts - a quick copy and paste has saved more than a few posts. Whether this is or is not a good thing…

Not quite:

  1. It’s completely legal to produce new human ESC lines. You just can’t do it, or any subsequent research using those new lines, with US government money. Of course, the majority of biological research in the US involves government grants, so…

  2. Research on somatic stem cells started first, and in some cases is further along. Human ESCs were first derived less than twenty years ago.

  3. If a patient has banked cord blood, it’s possible that they could be treated with their own cells (suitably manipulated), which would avoid the rejection issues that ESC-derived cells (which wouldn’t match the patient) would have. This is a big advantage, since graft rejection is a major unsolved problem for almost all forms of transplant therapy, stem-cell related or not.

Sometimes; sometimes the goal is to make the desired cell types directly. There are many avenues of research, to make different tissue types, to be useful for different purposes, etc.

There are many, many research groups interested in stem cell therapy, for a wide variety of diseases. But only a few possible therapies are anywhere near being ready for human trials, much less general clinical use. It depends on multiple factors; a couple of the most important are: whether we already know how to make the target cells; how close to the “real thing” the lab-made cells are; what kinds of manipulations are involved (some techniques require forcing the cells to express genes that have a potential to cause cancer, which of course is a big downside); and how easy it would be to use the final cells therapeutically.

The last one is often overlooked in popular reports, but it can be one of the hardest problems. For example, insulin-dependent diabetes is relatively easy to treat with transplanted cells: as long as you can get the insulin-making cells into a place where they have a good blood supply, they’ll be able to do their thing. But the kidney not only has many cell types but is a complicated three-dimensional structure; just stuffing cells in anywhere won’t do you any good, each cell type has to be in the exact right place along the length of each of the hundreds of thousands of tubules. Stem cell therapy for the kidney, if it ever happens, will require not just making the different kinds of kidney cells but assembling them into a complete artificial kidney in all its glory; we can do the former but we are nowhere near being able to do the latter.

Ok, you lost me on that one (not a difficult maneuver; don’t pat yourself on the back).
Can the kidney not repair itself via somatic stem cells? Is it common (as of current understanding) for an organ to be incapable of regenerating its tissue(s)?

Would it be possible to produce magic cells which could become any tissue, which could be inserted in such a manner as to cause them to become whatever “should” be there?
Inserted into a spinal cord, they begin to produce appropriate nerve cells? Injected into a damaged cornea, become cornea cells, etc? Is there a blueprint somewhere (DNA?) that could be invoked? Or just “look around and act like every other cell around you”?

Is this even remotely possible, or will never get beyond growing the (single) desired tissue and then inserting it in the exact place it needs to be?
Which, of course, is real close to miracle all by itself.

No, the kidney can repair itself, the necessary somatic stem cells are there, in the right places, but we can’t treat injuries to, or diseases, of a kidney with stem cell therapy, because it’s s complicated structure where everything has to be just in the right place.

In contrast with, to repeat all of the previous posters points, insulin producing cells which can just be stuffed in “anywhere”.

Most organs have some ability to repair / regenerate themselves, but at the same time most organs have a limit on how much damage they can take before their self-repair ability gives out. Some diseases, like liver failure, kidney failure, and type II diabetes, can be caused by mild but continually-recurring damage that eventually wears out the tissue’s ability to regenerate. In contrast, some systems have phenomenal regenerative ability; the entire blood supply can regenerated from a single hematopoetic stem cell, and you can knock it back to a single HSC over and over again and have it grow back fine every time.

This is almost certainly impossible, sorry; there are far too many types of cells. Some stem cell types move around to find places they are needed and make the right thing in the right place; for example, “vascular endothelial stem cells”, which make the lining of the blood vessels, will migrate towards areas that need more oxygen and build new vessels there. So for cells like that, getting them into the right general area is enough. But for many tissues, the mechanism for getting the right cells into the right places only exist in the embryo, and therapeutic use of stem cells will require figuring out not only how to make the cells but how to assemble them into the right architecture.

Again, thanks for all the thoughtful (and patient) responses.

If I understand the embryonic development sequence:
Step 1: Generic stem cells
(followed very, very closely by)
Step 2: Semi-specific stem cells (as in may produce any of several, but not close to unlimited cell types)
Step 3 Fully differentiated adult cells, with a few or the Step 2 type of stem cells retained for repair/regeneration.

I was hoping the magic cells could operate along the lines of immune response, but I neglected to note that dead/missing cells probably do not send out a “replace me” signal.

Here’s another off-the-wall thought: with all the study of the immune system function, there is/are a specific cells which detect an infection and direct the immune response (I want to say T-helper).
Could there be an analogous cell to detect dead/defective/missing tissues and trigger a regenerative response?

There’s no special class of cells that does this, but many ordinary cell types do it for themselves.

One mechanism is that cells talk to each other a lot, and there are many signals for “we need more over here”. Injured cells of certain types send out a generic “mayday” signal, and any kind of dying or damaged cell will leak its contents (cells are basically little bags of liquid, and the liquid inside the cells contains a lot of stuff not normally found on the outside); other “garbage collection” and “general maintenance” cells detect these things and come over to help clean up debris and corpses and to repair the “scaffolding” that cells stick to or crawl along (the extracellular matrix). I mentioned that blood-vessel stem cells migrate towards regions that aren’t getting enough oxygen; they don’t detect low oxygen directly, but signals that are sent out by oxygen-hungry cells. Some cells, like skin cells, want to have neighbors on every side; if they detect empty areas around them, that triggers growth and migration to cover the bare spots. There are many mechanisms like this, but most of them are specific to the tissue in question; no one cell has the capacity to detect every possible problem and send out the corresponding signal.

Another is that, in many tissues, cells have a way of measuring how hard they are working and will trigger the system to make more cells if they are working too hard; under normal situations this helps match demand (if you exercise, you will get not just bigger muscles but actually more muscle cells), and after injury it helps regenerate missing cells. Not all tissues can do this, and even in the ones that can, the amount of regeneration may eventually be exhausted, especially if the tissue has to regenerate repeatedly (as in many chronic diseases).

People are interested in using both kinds of mechanisms for therapy; for example, there are treatments under investigation that try to apply the “more blood vessels please” signal to treat heart disease. As usual, getting the signal to the right place is the hard part; you don’t just want a bunch of extra blood vessels everywhere…

Where are those magic wands and why do they always disappear when needed most?

I just read the story (linked in MPSIMS for all the wrong reasons) of successful creation and implantation of lab-grown vaginas. Story here

The part most interesting: at least from that write-up was that the vulva tissue was convinced to produce sheet of tissue which was mechanically formed into a tube and implanted - and then it grew its own blood supply (not too surprising) and nerve cells. Didn’t we used to know that neurons could not be regenerated (hence spinal damage is permanent)?

I’m guessing the cells used were somatic stem cells. And the patients were still teenagers, so normal growth was still going on.

Are we really closing in on re-generating neurons? (yes, it is on my “google this” list).
Could this process be expected to work as well in adults as in children/teenagers?

I took a quick look at the report in Lancet and although they don’t discuss it in detail, what I think the authors are saying is that the implanted tissue was able to connect with nerve cells in the area, not that brand-new nerve cells arose in it. Neurons have to physically touch the cells they connect with, so they grow long “fingers” to reach out to their targets (up to a yard long, in some cases); I think that in this case the transplanted tissue was able to convince nearby neurons to grow “fingers” in its direction and connect to it.

In any case, some types of neurons in the body can regenerate; it’s the ones in the brain and spinal cord that we thought were irreplaceable. However, some years back we discovered that there are actually a few areas of the brain that grow new neurons; it’s restricted to specific cell types in small regions, and a relatively small number of cells, but it does happen. People are very interested in inducing regeneration in the other parts of the brain too, but that is far off (if it ever works).

The vagina (and the other mucous membranes, like the mouth lining) is a continuously renewing tissue, like skin, so it has stem cells that never go away; vaginal cells from younger individuals can grow more robustly, but even in very old women vaginal cells can still grow. So it should also work with adult cells, although probably a little more slowly.