Re: do you think it may be possible, that with the use of genetic engineering

"I'd give my right arm to know the secret of regeneration."

- Oscar Schotté {1}

(Note: This is a long answer. The short version is: "Probably, but it ain't
easy, and it's a lot simpler when there's already some regenerative
machinery in place. We'll probably be able to fix spinal cord injuries
soon, and some organ damage, but replacing an arm is hard.")

First, we have to be careful to define the kind of regeneration we're
talking about. You mention several types; let's start with what may be the
toughest of the lot. I'll dwell on it because it's particularly tantalizing
and difficult.

It's called epimorphosis, and it's the reason why salamanders can regrow
lost limbs. Epimorphosis is regeneration by cell dedifferentiation and
redifferentiation. In other words, cells in, say, an amputated leg stump
lose their identities, turning from skin, muscle, bone, and nerve into a
blob of cells called a blastema. This blastema has some resemblance to the
cells of a developing embryo, so your intuition that regeneration would
require us to "simulate" fetal growth is on the right track. {2} These
cells then grow and are patterned into the lost limb in a process that is
still mysterious, but does, at some points, strongly resemble the limb's
initial development.

Is this possible in humans? We know that if you cut off the fingertips of
mice, they can regenerate. Surprisingly, this is also true in human
children! But the process doesn't look like the salamander regeneration.
{3,4}, as there's no blastema {4}, and it won't get you a new arm.

Part of the problem is that that the cells making the regenerated limb have
to somehow set up a proximal-distal axis. In other words, they need to
"know" to turn into fingers at the end of the limb, into wrist closer to
the body, into forearm when still closer, and so on. In both the salamander
and mammals, when the fetal arm is growing, genes called Hox genes help
specify this; different ones are active in different parts of the arm. But
in the salamander, Hox genes stay active throughout adulthood. Not so in
us. {2}

This may explain why, if you cut off a salamander's leg, then take cells
from the stump and grow them elsewhere on the salamander, a leg forms, but
if you cut off its hand, and stick the cells from that stump elsewhere,
they "remember" that they used to be wrist cells, and become a hand. We
can't pull the same trick, of course. This cell memory definitely isn't
SOLELY due to the Hox patterning; it likely has much to do with the
expression of other 'markers' of fate. But that's probably involved, and,
whatever the memory marker is, we don't have it. {4}

(And yes, you can grow a salamander arm on other parts of its body. In its
eye, for example.)

But suppose we could artificially generate a mess of dedifferentiated cells
on a stump (see below), somehow fit a protective sheath over it, and try to
remake a pattern of Hox activity. And we mustn't forget all the other
relevant location signals, either, since we don't want to end up with, say,
thumbs instead of pinky fingers (an anterior-posterior patterning defect)
or palms on both sides of our hands (a dorsal-ventral defect.) It's
complicated, but could it work?

It's a long-shot, but conceivable, far in the future. This is where stem
cells come in handy.

Stem cells have a lot in common with a newt amputee's blastema. (Not just
embryonic stem cells, by the way, though those are especially useful.) They
are pluripotent, i. e., they can become many cell types, and they can still
divide, unlike fully differentiated cells. They could theoretically be
implanted in tissue to generate multiple cell types. It's worth a shot, and
there has been work in mouse models. But an arm - or even a finger - is
complicated, and we don't know enough about how one is made yet, or what we
would do if we had the knowledge. I could be wrong, but I think we're a bit
of a way off. I wouldn't want to give myself a paper-cutter manicure
anytime soon.

Now for the other kinds of regeneration - the kinds that would apply to
damaged organs, burned skin, and the neurons in a broken back. The problems
are still tough, but the prognosis is much more hopeful. We don't have to
make a totally undifferentiated blastema mass here. One strategy is to give
the cells that are near the injury some "blastema-like" properties - let
them regain their ability to divide and replace the injured tissue. In
other words, we either create or implant stem cells, and let them take
advantage of the extracellular scaffolding that's already there to guide
them into place and help them fit in.

In some cases, we might not even have to do that much; we can stimulate the
growth of cells that are already present to replace the dead. Some organs
can do this without our help; the liver, for example, can regrow if you cut
pieces of it off, hence Prometheus's mythological liver-removal torture.
Likewise, we heal cuts in our skin. So our simplest goal would be to
enhance regeneration processes where they already naturally occur on some
level. This is why some of the biggest success stories so far have been
treatments that cause the axons of neurons in the rodent spinal cord to
re-grow. But even here, stem cells have been helpful to coax healing along.
 We have a long way to go. {5}

But even here, we don't quite understand enough to pull this off with
certainty, though preliminary trials are looking good. In some cases,
careful tinkering can get regeneration to occur, but we don't know exactly
WHY it works. Around the world, labs are studying differentiation,
dedifferentiation, and regeneration in a gamut of systems, in organ tissue
culture, mice, newts, frogs, and even flatworms, trying to work this out.
We have a very long way to go.

It was only last year that Peter Reddien (MIT) and Alejandro Sánchez
Alvarado (University of Utah) figured out how to systematically find genes
that let _flatworms_ regenerate. And they had to do it by basically
knocking down one gene at a time until they saw flatworms that didn't
regenerate. We have a very, very long way to go.

In the lab where I work, some researchers are interested in figuring out if
we can replace or regenerate the insulin-producing beta cells of diabetics.
That's not making a new organ. That's just making one kind (or a few kinds)
of cell. They actually regrow naturally under some circumstances. But it's
still a hard problem.

Did I mention that we have a long way to go?

We have at least some of the tools to get the info we need - genome
sequences, stem cell lines, tricks for looking at gene expression to tell
us how regeneration is programmed - but we need to create new ones. This is
a big, open field - and the funding is excellent.

You choose your interests well...

Paul Nagami

P. S. If you're interested, I've got some references below. Also, try
Googling the researchers I mentioned, and searching for others. _We need
seriously need more scientists to deal with this._

{1} Quoted in Developmental Biology, 7th ed. (2003) Scott F. Gilbert.

{2} Gilbert, Scott F. (2003) Developmental Biology, 7th Ed. Sinauer Press

{3} Muller, T. L. et al (1999) Regeneration in higher vertebrates: Limb
buds and digit tips. Seminars in Cell and Developmental Biology,
10:405-413. Available here: http://cell.tulane.edu/Muneoka/Publications/muller%20et%20al%2099%20SemDB.pdf#search=%22fingertip%20regenerate%22

(^ This one has nice, if fairly gruesome, pictures, and a pretty
straightforward description.)

{4} Brockes, JP and Kumar, A. (2005) Appendage Regeneration in Adult
Vertebrates and Implications for Regenerative Medicine. Science 310:1919-1923

(A good broad review. Available here, if you can access it for free: http://www.sciencemag.org/cgi/content/full/310/5756/1919)

{5} Lindvall, O and Kokaia, Z (2006) Stem cells for the treatment of
neurological disorders. Nature 441:1094-1096