THE water was cool and still, perfect for a dip. The swimmer had struck out
confidently from the shore, when suddenly a dark shape appeared below him. There
was a commotion in the water, and then a stabbing pain in his left leg as a set
of sharp teeth sank into his flesh. Thrashing his body from side to side in
panic, he wrenched himself free, then struggled the short distance to shore. But
his leg was still in the mouth of the cruel predator.
Unfazed, the swimmer took things easy for a few weeks, and sure enough his
patience was rewarded鈥攈e grew a brand new leg. A fairy story? Far from it.
Our swimmer is a newt, and regeneration comes naturally. But perhaps the same
story could soon come true for humans, too.
There are good evolutionary reasons we mammals traded in the ability to
regenerate damaged tissues鈥攁n advanced immune system and rapid wound
healing among them. But new work is showing that a newt-like capacity for
rebuilding whole limbs might still be locked within our cells. Some researchers
even believe they鈥檙e close to working out how to set the chain of events in
motion to allow the growth of a whole new arm or leg.
Advertisement
The secret of the newt鈥檚 ability to regenerate body parts lies in ordering
specialised cells at the wound site to revert to an embryonic state, and using
these to build a new limb from scratch. Once a cell has become a bone cell or a
muscle cell, it usually can鈥檛 convert into other types of tissue, and often
can鈥檛 divide. But by reversing their development, the newt turns its cells into
blank slates again, forming a ball of precursor cells called a blastema, which
is able to grow, multiply and re-specialise to make any of the tissues needed.
In mammals, this regression simply doesn鈥檛 happen. Instead, the wound heals
(see Diagram). It鈥檚
a process that has evolved to close the wound as fast as
possible to minimise the chances of infection, says Mark Ferguson, head of the
Wound Healing Research Group at the University of Manchester. But speed often
means leaving a scar.
Dan Neufeld from the University of South Dakota School of Medicine believes
wound healing and regeneration are competing processes, with the balance in
humans tipped towards healing. 鈥淭hey race against each other,鈥� he says. 鈥淚
believe that in humans, regeneration is an inherent ability. But we have such
fast and effective wound healing, it always gets in first.鈥�
Healing is the job of the immune system, and animals that regenerate have
relatively primitive immune systems. Even mammalian embryos, before their immune
systems are fully developed, can heal without scarring. Neufeld and others hope
that since we can now dress wounds and control infection, we could slow down
healing to give regeneration a chance.
It looks as though this might be possible in mice, as Ellen Heber-Katz, an
immunologist at the Wistar Institute in Philadelphia, discovered. She was
working with a strain of immune-deficient mice as a model for the human disease
lupus and got a student to punch tiny holes in the ears of the mice鈥攁
well-established technique for tagging the animals. But three weeks later, the
holes were gone.
The holes hadn鈥檛 just healed. As well as forming new skin the mice made new
hair follicles and cartilage. And the immune-deficient mice could also regrow a
lost tail tip, making new bone, cartilage, skin and hair. Heber-Katz saw that
the mice were forming blastemas, just as newts do. 鈥淭his is something that is
essentially never seen in mammals,鈥� she says. 鈥淚t looks like limb
谤别驳别苍别谤补迟颈辞苍.鈥�
Hidden talents
She recently teamed up with Elizabeth Blankenhorn, a geneticist from the
Medical College of Pennsylvania, to find out which genes are responsible for
immune suppression and blastema formation in the mice. They found that many of
the genes that are active during limb regeneration in newts are also found in
these mice, so it really does look as though mammals could have hidden
regenerative talents. 鈥淎lmost everything seen in amphibians, we see here,鈥� says
Heber-Katz.
But immune suppression is clearly not the whole story. The regenerative
capacity of the mice is by no means perfect. 鈥淭he bone seems to grow faster than
the rest of the tissue,鈥� says Heber-Katz. And repairing tail tips and pierced
ears is a far cry from growing whole new limbs.
Ferguson and his Manchester colleagues are now studying the differences
between adult and embryonic immune systems and healing mechanisms in the hope of
learning what makes an immature immune system allow perfect healing rather than
scarring. One of the main differences they have found so far is in the levels of
a family of chemicals that promote growth. Embryos have lower levels of two
growth factors called TGFb1 and TGFb2, but more TGFb3.
The researchers altered the balance of these growth factors in adult mice,
rats and pigs to match the embryonic levels, then cut out little pieces of skin.
They found that when the animals healed the tissue didn鈥檛 scar. 鈥淭he wounds heal
without a mark,鈥� says Ferguson. The group is now carrying out preliminary
clinical trials in humans, and should have the results next year. But even
though the wounds healed neatly, they did so only with new skin cells鈥攏o
new sweat glands or hair follicles were formed. The new tissue was generated
through the division of skin cells around the wound. There was no regression and
redifferentiation, so no other tissue types were made.
Now Jeremy Brockes and Elly Tanaka from University College London are trying
to work out what it is about wounding that makes mature newt cells, but not
mammalian ones, return to their embryonic state. They are studying newt muscle
cells, which fuse into aggregates called myotubes. When the cells begin to
regress, the first step is to break up again into their individual units. And
the signal to do this seems to be exposure to blood serum. Brockes and Tanaka
think the key might be thrombin, an enzyme involved in clotting
(快猫短视频, 31 July 1999, p 10).
The precise details of what thrombin does aren鈥檛 clear, but the endpoint
seems to be the modification of a protein called retinoblastoma, or Rb. In
mammals, Rb is an important safeguard against cells dividing recklessly, and so
is important in protecting against cancer. But when Rb is modified by a
complicated cascade of reactions triggered by thrombin, it releases its grip,
allowing the cells to regenerate.
The researchers were surprised to find that the thrombin signal to start
regenerating isn鈥檛 species-specific. Blood serum from mammals can trigger newt
myotubes into regressing, although mammalian cells don鈥檛 respond to the signal.
But when Brockes and Tanaka made hybrid cells with a mammalian nucleus inside a
newt cell, hybrid myotubes made of such cells were triggered by serum to
regress, and their mammalian DNA was able to replicate. Rather than being
completely incapable of regeneration, it looks as though somewhere in mammalian
cells there is an obstacle blocking the pathway.
Peter Schultz from the Scripps Research Institute in San Diego took a brute
force approach to overcoming this obstacle. One by one, he added thousands of
different chemical compounds to mouse myotubes. He found one that induced
myotubes to break into their individual units. And when he removed the
substance, which he called myoseverin, the cells started to replicate. 鈥淚t鈥檚 a
remarkable finding,鈥� says Brockes. Myoseverin causes the fragmentation by
dismantling tube-like structures within the cells. But that鈥檚 not all it does.
Schultz found that myoseverin also turns on a wide range of genes, many of which
are known to be involved in healing and regeneration. Starting the cells off on
the path towards an embryonic state seems to trigger genetic changes that could
be involved in tissue repair.
But simply knowing that it鈥檚 time to make a new part isn鈥檛 enough. You also
have to know what to make. Newts always know whether to make a leg or an arm,
and how much to grow鈥攁 single digit or a whole limb.
Hans-Georg Simon, a paediatrician at the Northwestern University Medical
School in Chicago, has identified a group of genes called T box genes, some of
which are active only in the forelimb, and some only in the hindlimb. Simon has
found that in mammals and other animals that cannot regenerate, these genes are
working in the limbs during development, but are then shut off completely.
In newts and other regenerating animals, on the other hand, the genes are
active at a low level throughout adult life. A kind of memory, or map, pervades
the newt鈥檚 body and gives each part its identity. 鈥淚f there is amputation, there
is always a background that tells the stump, `you have to make an arm鈥�,鈥� says
Simon. 鈥淚n us, these genes are completely shut off.鈥�
Simon believes this information is critical. In humans, mutations in a T box
gene that is switched on in arms during development cause a disease called
Holt-Oram syndrome. The symptoms vary, but some patients have no arms. It seems
that even if the signal to grow a new limb is there, says Simon, if there is no
information about what to make then you don鈥檛 make a limb at all. If we want to
regenerate limbs in mammals, we need to find a way to switch these markers back
on.
It may sound as though far too many signals have yet to be found to consider
human limb regeneration even a remote possibility. But Doros Platika, head of
the Boston-based company Ontogeny, is optimistic that one day we will be able to
trigger regeneration in mammals, whether we understand all the details or not.
He believes that our cells retain an inherent capacity for building new parts,
just as they did during development. The instructions for building each limb are
written in the DNA of every one of our cells鈥攖hat鈥檚 how we knew how to
make them in the first place. If we can just work out how to turn the ability
on, Platika believes the rest of the process will take care of itself.
Ontogeny has teamed up with another company called Stryker Biotech to develop
one of the first genetically engineered regenerative drugs, which they hope will
be approved for sale in Europe, the US and Australia by the end of the year. It
is a growth factor called OP1, and it stimulates new bone growth. OP1 is used to
treat non-union fractures, where a patient has broken a bone and there鈥檚 a gap
between the two pieces, so they don鈥檛 fuse and heal properly. Often this means a
limb has to be amputated, but OP1 stimulates bone to grow across the gap and
bridge the two pieces. It triggers a cascade of events which ensure that the
bone is fed by blood vessels, grows in the correct shape, fuses with the old
bone, and stops when the process is complete.
鈥淓ven if we keep on giving the drug, the response is controlled, and it stops
at the right time,鈥� says Platika. All doctors need to do is give the signal that
says to the bone cells 鈥済row鈥濃�攖he body knows how much bone to make and
where. Platika believes that if we can go back one step earlier in the process,
and find the growth signals for all the necessary cell types, then growing a new
leg really will involve nothing more than a few injections.
But even if we could make limb development happen all over again, there are a
couple of catches. One worry is that overriding normal safeguards and
encouraging our cells to proliferate could trigger cancer. Animals that can
regenerate seem to be particularly immune to cancer. When the control systems
for cell division fail, the animals grow new body parts rather than tumours. But
it鈥檚 hard to tell what would happen in mammalian cells.
Another catch is that when we developed our limbs first time round as
embryos, the chemicals that dictate the shape of the new limb could easily
diffuse from one side of the tiny body to the other. In adults, the distances
involved are considerably larger. We could solve the problem by forming a
perfect limb on a very small scale, then grow it up. This is exactly what newts
do. But how long would a whole leg take to grow to full size?
Newts can rustle up a limb in a couple of months, only we鈥檙e a lot bigger
than newts. Brockes worries that even if we could grow new limbs, the technique
would take far too long. It took around 18 years for our legs to get to their
current size, and regenerating a new leg won鈥檛 be very helpful if we have to
wait that long.
Nevertheless, Neufeld is optimistic. He points out that it doesn鈥檛 take a
newt much longer to regenerate a whole limb than it does to regenerate a hand.
鈥淥nce you have a certain number of cells, it just takes one or two more cell
cycles to make up the extra. My guess is that the scale of time in humans would
be roughly comparable to that of newts.鈥� Platika agrees: 鈥淭his is pure
speculation, but I see no reason why we couldn鈥檛 grow a leg in weeks or
尘辞苍迟丑蝉.鈥�
So how long will it be before doctors can perform this amazing feat? Platika
reckons five years. 鈥淚t鈥檚 an aggressive estimate,鈥� he admits. 鈥淏ut if you鈥檇 told
people five years ago that we鈥檇 be able to clone sheep and cows, they wouldn鈥檛
have believed that either.鈥�
People can sometimes regenerate the tips of fingers and toes, so long as the
wound runs through the nail bed, which contains a supply a stem cells. Maybe if
we can tap into natural supplies of these precursor cells, we could avoid the
need to send mature cells back into their embryonic state before regeneration
can happen.
Dan Neufeld and Khalid Mohammad from the University of South Dakota School of
Medicine have been studying the process in rats. They transplanted the nail and
its bed from an amputated toe tip to an amputation site higher up the limb.
Normally a stump without a nail bed would just heal at that position, but
instead the nail stimulated the regeneration of a normal-looking toe. 鈥淣ot only
was something induced,鈥� says Neufeld, 鈥渋t was re-establishing a functional
anatomy.鈥� But it didn鈥檛 manage to regenerate a whole leg. 鈥淢aybe the toe ran out
of tissue,鈥� Neufeld suggests.
He鈥檚 not certain, however, whether this proves that stem cells are the key to
regeneration. The role of the nail bed could be to provide stem cells or growth
factors that trigger regeneration. Alternatively, it could simply be a matter of
geometry. 鈥淭he critical factor could be that the nail holds the wound open,鈥� he
says, suppressing the healing process and giving regeneration a
chance.