More Stem Cell Magic

How long will it be before we have replacement tissues made from in induced pluripotent stem cells (iPS cells, iPSCs, or sometimes iPSs)?  Progress in generating iPS cells in university labs has been rapid, resulting in a series of recipes that are spreading capability rapidly around the world.  About 18 months ago, in the wake of progress on creating iPSs from adult, differentiated tissues, I started wondering about how long it would take before someone tried this in a garage (see "Stem_Cells@Home or DIYStemCells?").

Accessible Is Not The Same As Easy

In what follows, please keep in mind that I am not asserting that producing autologous iPS cells will be easy for anyone.  It will be hard.  And it will be harder for those attempting the feat in settings like garages and small start-ups.  However, I visited a garage lab last spring in the SF Bay area that was more than adequately equipped to give producing iPS cells a go.  As the highly technical protocols become recipes, more people will have the opportunity to try them out.  This was my point in Carl Zimmer's piece in the New York Times last week.  Innovation requires the opportunity to fail. 

On the Path to StemCells@Home

This story starts, for me, with the demonstration in June of 2009 that recombinant proteins can be used to reprogram skin cells into iPSCs (see "Another Step Toward DIYStemCells").  Previously, this reprogramming step required genetic manipulation via viruses, which greatly lessened the utility of the resulting iPS cells for therapies.  With the demonstration that proteins themselves could be used to reprogram cells, anyone who wanted to spend ~$10K on synthesizing four genes and then another ~$10K on having the four corresponding proteins made in cell culture could have those proteins delivered by post a few months after the initial order.

As I commented last year when these results were announced (see "Another step toward DIYStemCells"), "if you wanted to do this at home, you could.  You should expect to fail many times.  And then you should expect to fail some more.  And then, assuming your human cell culture technique is up to snuff, you should expect to eventually succeed."  That is just the way it works in university and corporate labs.

But wait, it is now even easier to make iPS cells!  In September, a paper from researchers at Harvard showed how to use RNA to reprogram adult cells into iPSCs.  Writing in the Washington Post, Rob Stein described the advantages of the new method: "The technique converted the cells in about half the time that previous methods did, about 17 days, and with surprising economy - up to 100 times more efficient."

Here is the Warren et al paper at Cell Stem Cell: "Highly Efficient Reprogramming to Pluripotency and Directed Differentiation of Human Cells with Synthetic Modified mRNA".  The team used a combination of chemical modifications of RNA, along with packaging of the mRNA in cationic lipids, to reprogram a variety of differentiated cells into "RiPS cells".  Most of the "chemical modifications" consist of changing a standard RNA synthesis recipe to include non-standard ribonucleotides, followed by a bit of enzymatic trickery.  The authors then used the same RNA programming trick to control the differentiation of those RiPS cells into a variety of different tissues:  "Our results demonstrate that modified RNA-derived iPSC clones from multiple independent derivations were fully reprogrammed to pluripotency and that the resulting cells very closely recapitulated the functional and molecular properties of human [embryonic stem cells]."

There are a few "Technical Notes" at the end of the paper.  Warren et all recommend that "all steps of the protocols described herein are followed rigorously and quality controlled."  They also observe that "Critically, the expression of proteins with modified RNAs must be confirmed by immunostaining."  Basically, this recipe sounds finicky and probably requires a great deal of practice.  (To say that I am oversimplifying here is to say that Hurricane Katrina was a wet sneeze.)  But oh what an improvement it is over the prior methods for making iPSCs.

Rob Stein, at the Washington Post, and Karen Weintraub writing in Technology Review, describe how the entire Harvard Stem Cell Institute is going to start using Rossi's recipe to make iPSCs, and how researchers at other institutions plan to try it out as soon as they can.

Although it is relatively technically complex, the methodology described here offers several key advantages over established reprogramming techniques. By obviating the need to perform experiments under the stringent biological containment required for virus-based approaches, modified RNA technology should make reprogramming accessible to a wider community of researchers.

In other words, Warren et al published a recipe.  A complicated, recipe, to be sure, but a recipe that is already being used (and probably improved) in a large number of labs around the world.  Does that mean we will see autologous stem cell transplants next year?  Probably not.  But we might.  The FDA, please recall, only regulates drugs and devices [thanks for the reminder, Bill], and only then through the Interstate Commerce Clause of the US Constitution.  The FDA is explicitly prohibited from regulating treatments, which are designed and implemented by doctors.  So as long as stem cells are used in procedures considered therapies, the FDA doesn't have anything to say about the use of RiPS cells in patients.

And regardless of progress in the clinic, at some point this technology is going to be tried by "the wider community" in the garage.  It is inevitable.  And when a garage protocol is successfully demonstrated, and perhaps shared in among people participating in Open Biology, then we will see a profusion of new therapies.  And also a profusion of mistakes and strange teratomas, because iPS cells will be used in contexts where nobody has any idea what the consequences will be.  But that is also inevitable.  Once producing stem cells truly moves from art to recipe, I don't think there is any way to stop people from playing with their own stem cells. 

So it is Magic, or Science?

Actually, it is starting to look a lot like engineering, or maybe even cooking.  By "magic", I mean not Harry Potter but art, or something that nobody really understands and works only in that hands of a small number of people.  "Science" in this context would be experiments that are designed to test particular hypothesis or to develop new methods, in both cases resulting in descriptions of nature or methods that require substantial reduction to practice before adoption is widespread.  But the RiPS method looks like it is being implemented widely just weeks after publication.  Nobody fully understands why RNA reprogramming works, or how RNA-directed differentiation works, to be sure, but this method is suddenly much closer to an engineering protocol than a mysterious incantation that only a few artists can implement.

Organs@Home or DIYOrgans.  More Likely DIYTumors.

So what are we going to use RiPS cells for?  Reaching back to news over the last year points the way.

Rob Stein, again writing in the Washington Post, described in July 2009 how mice were grown from iPS cells made from adult skin cells.  And in a news piece at Nature, David Cyranoski elaborates on the efficiency of the process as well as how many additional generations of mice were grown from the initial litter.  We've also now seen replacement teeth grown from stem cells (WSJ).  (Here is the Ikeda et al paper in PNAS: "Fully functional bioengineered tooth replacement as an organ replacement therapy".) 

In July of this year, the same sort of viral hack was used to make iPS cells from leukocyctes found in adult peripheral blood samples.  Laura Sanders at ScienceNews described the papers succinctly: "Blood drawn with a simple needle stick can be coaxed into producing stem cells that may have the ability to form any type of tissue in the body, three independent papers report in the July 2 Cell Stem Cell."

Shinya Yamanaka (who originally demonstrated the use of the 4 "Yamanaka factors" in producing iPS cells) wrote a very clear commentary accompanying the three papers.  First, here are links to the three papers: Seki, et al; Loh, et al; Staerk, et al.  The last paragraph of Yamanaka's piece is full of cautions about the utility of iPSCs derived from peripheral blood.  In particular, Yamanaka notes that his group showed that the safety of iPSCs in mice depends on the origin of the tissue used to generate the stem cells.

There are many hurdles to overcome before iPSCs are used in the clinic.  But the Seki paper in particular shows generation of stem cells from the T lymphocytes in just 1 ml of blood (becoming "TiPS cells").  That 1 ml of blood was put through a relatively straightforward Ficoll separation column to enrich the sample for T lymphocytes.  Obtaining these cells is pretty simple, and is in fact something I did myself, using my own blood, many years ago for an experiment for my doctoral work.  And most of those experiments were done with only a few hundred microliters of blood extracted from a finger stick.  In other words, I can imagine at least starting down the road surveyed by Seki et al with just a lancet (a sterile needle would do), a Microtainer with heparin from BD, and microcentrifuge.  In fact, here is a protocol from Ohio State (PDF) that looks like it would do fine to derive your own T cells, though you could probably skip the red blood cell lysing step.

The initial separation is followed by culturing the T cells in a dish, which also isn't so hard as long as you have the proper equipment.  However, thereafter the Seki et al recipe starts to get a little hairy, including multiple steps of culturing on feeder cells and incubation with very specific kinds of cell extracts.  If you try this in your garage, you are likely to fail many times.  But that is to be expected, because Seki et al failed many times, too.

Seki suggests one reason for the low efficiency of conversion T lymphocytes to iPSCs is due to the low rate of gene transfer by viruses.  Now, presumably, you see where this is going:  Re-enter Warren et al and their RNA induced pluripotent stem cell method described above.  Recall that this method works in about two weeks and is ~100 times more efficient in generating iPSCs than is gene transfer.  No doubt there will be some hurdles to overcome before putting all these pieces together, but I would be greatly surprised if there we didn't see RiPS generated from adult peripheral blood cells by the middle of next year.  Undoubtedly that paper will also demonstrate some streamlining of the protocol.  And then people will have another recipe to play with.

Seki et al demonstrated that human TiPS cells implanted into immune compromised mice can differentiate into many different tissue types.  They also showed that the TiPS cells can become teratomas, which means unless you are careful with the implantation of these cells you are going to wind up with strange tumors.

That said, Warren et al show that subsequent RNA reprogramming can direct RiPS cells to become all sorts of interesting tissues.  So if you want to try all this in your garage, and if you have the appropriate cell culture skills and equipment, you can give it a fair go.  Access to the appropriate strains of feeder cells, as well as the modified RNAs, could be a stumbling block.  But I have to imagine 1) that those cells and the RNAs are going to be available commercially as a package at some point or 2) that you will be able to get the cells from a supplier and contract out the RNA production for no more than a few thousand dollars.

Derek Rossi, the Harvard professor who is the senior author on the Warren paper has, according to the acknowledgments on the paper, started a company "dedicated to the clinical translation of this technology".  Whether that means there will be a monopoly on the methods and materials is unclear to me at the moment.  If you want to generate your own RiPS cells from T lymphocytes, who is going to stop you?  And if you use those cells to produce tissues, and even to attempt treating yourself?  Even then, it isn't clear that there is any rule, law, or regulation that can be used to stop you; recall that the FDA, at present, has not yet decided to try to regulate stem cells as drugs.  Please note that I don't think self treatment is a very good idea, just yet.  But nobody who is interested in playing with these technologies is likely to listen to me on this point anyway.

However, if you did something that looked like treating another person, then all hell would probably break loose because you could be accused of practicing medicine without a license.  And then there are the consequences of getting this wrong, whether you are treating yourself of somebody else.  About a year ago, on an airplane, I happen to sit next to the CEO of one of the largest health insurance companies in the US.  At one point in the conversation, I asked him what his company would do if people started showing up needing treatment for tumors they gave themselves by injecting their own iPS cells.  He just stared at me, stunned, with his jaw agape.

That is the right response, I suspect.  The world is changing very quickly, and even if you spend your days trying to understand what is coming you are guaranteed many surprises that will just leave your jaw agape.