杏吧原创

If you put your ear to the tracks, you can hear the train coming.

In conference halls around the world, geneticists and developmental
biologists have been gathering to discuss what once was
unthinkable鈥攇enetically engineering human embryos so that they, and their
children, and their children鈥檚 children, are irrevocably changed. These experts
are talking with remarkable candour about using germ-line engineering to cure
fatal diseases or even to create designer babies that will be stronger, smarter,
or more resistant to infections.

Doctors are already experimenting with gene therapy, in which a relatively
small number of cells鈥攊n the lungs, say鈥攁re altered to correct a
disease. Germ-line engineering, however, would change every cell in the body.
People would no longer have to make do with haphazard combinations of their
parent鈥檚 genes. Instead, genetic engineers could eliminate defective genes,
change existing ones or even add a few extra. Humanity would, in effect, take
control of its own evolution.

So awesome is this idea, that until a year or so ago, the taboo on human
germ-line engineering was absolute. But opinions have started to shift. Once
barely considered a topic for polite conversation among even the most gung-ho of
geneticists, germ-line engineering of humans is becoming so much grist to the
mill of scientists gossiping around the coffee pot.

Not that the pillars of the scientific establishment agree on this emerging
technology, not by a long way. In a straw poll, researchers variously described
the idea of human germ-line engineering as 鈥渋rresistible鈥�, 鈥渕orally
questionable鈥� or just plain 鈥渄angerous鈥�. What they did agree on is that
germ-line engineered humans are likely to become a reality. Tampering with a
human embryo to create changes that can be passed from one generation to the
next is still more or less verboten鈥�23 countries have signed a
Council of Europe convention that bans it, and officials at the US Food and Drug
Administration promise not to give the go-ahead without much public
deliberation. Despite this, however, most experts say they鈥檇 be surprised if
designer babies are not toddling around within the next 20 years or so.

Gregory Stock, a biophysicist-turned-expert on technology and society at the
University of California, Los Angeles, helped to organise a symposium in March
called 鈥淓ngineering the Human Germline鈥�. The task? Not to look way into the
future, but at what we鈥檒l be faced with in the next decade or two. 鈥淭here is no
way to avoid this technology,鈥� explains Stock, who thinks that calling the
evolutionary shots will create a happier, healthier society. 鈥淭he knowledge is
coming too fast, and the possibilities are too exciting.鈥�

Public enthusiasm could soon match Stock鈥檚: poll after poll shows that a
sizeable minority of parents鈥攕ometimes as many as 20 per cent鈥攕ay
that they see nothing wrong with genetically altering their children for health
reasons, to give them an edge over the child at the next desk鈥攐r even to
stop them being homosexual.

So what is shifting the mind-set about human germ-line engineering from
鈥渘ever鈥� to 鈥渨ell, maybe鈥�? The main driving force, most experts agree, is the new
technologies rolling inexorably along the tracks. We are discovering not only
what our genes do, but how to make precise changes in them. And although the
human genome isn鈥檛 yet completely sequenced, already the databases contain
details of thousands of genes, and of thousands of variations within them, along
with information about how these variations affect physical and emotional
traits. Added incentive comes, paradoxically, from frustrations with gene
therapy.

Gene therapy promised to cure genetic disorders such as cystic fibrosis and
sickle-cell anaemia, and even common illnesses such as cancer. But although the
glitches are slowly being fixed, few people have so far benefited from the
procedure. The problem is getting new genes into enough cells, and keeping
them there for long enough to do any good.

With germ-line engineering you have to tweak only one cell鈥攁 fertilised
human egg鈥攚hich is 鈥渋nfinitely easier鈥�, says Leroy Hood, a molecular
biologist at the University of Washington in Seattle. 鈥淲e have terrific ways to
do 迟丑补迟.鈥� Once a genetic engineer has changed the genome of an egg fertilised in
a lab dish, the egg divides over and over again, forming all the tissues of the
body. Every cell will have exactly the same genetic make-up as the altered
egg.

Now and forever

Genetically engineered mice and farm animals have been around for years and
are used for everything from basic research to attempts to create 鈥渉umanised鈥�
animal organs for transplant. But what might be considered a bonus in
agricultural biotechnology鈥攖he fact that any changes are present in the
animal鈥檚 sperm and eggs (the 鈥済erm cells鈥�) and so will be passed on to
succeeding generations鈥攊s for many the most worrying thing about genetic
engineering in humans. The critics point out that if medicine has played a bit
part in our recent evolution鈥攁ntibiotics, for example, allow people with
less than robust immune systems to survive long enough to pass this trait into
the next generation鈥攇enetic engineering has the potential to be a star
performer.

One reason for cold feet is that large-scale genetic engineering could
actually rob society of desirable traits. What if the 鈥渄isease鈥� genes in
combination with other genes, or in people who are merely carriers, also help
produce such intangibles as artistic creativity or a razor-sharp wit or the
ability to wiggle ones ears? Wipe out the gene, and you risk losing those traits
too. And while no one would wish manic depression on anyone, society might be
the poorer without the inventiveness that many psychologists believe is part and
parcel of the disorder. In his book Remaking Eden, Lee Silver, a
biologist at Princeton University, goes as far as to suggest that a century or
two of widespread engineering might even create a new species of human, no
longer willing or able to mate with its 鈥済ene poor鈥� relations
(鈥淯s and them鈥�, New 杏吧原创, 9 May, p 36).

鈥淭he potential power of genetic engineering is far greater than that of
splitting the atom, and it could be every bit as dangerous to society,鈥� says
Liebe Cavalieri, a molecular biologist at the State University of New York in
Purchase. Cavalieri, who has worked in the field for more than 30 years, thinks
it unlikely that the ugly side of genetic engineering will stop development of
the technology in its tracks. 鈥淚t is virtually inevitable it will get used and
for the most banal reasons possible鈥攖o make some money, or to satisfy the
virtuoso scientists who created the technology.鈥�

If esoteric worries about what may or may not happen in a genetically
engineered society are unlikely to change people鈥檚 views, safety issues
could鈥攁t least until they are solved. 鈥淭here is a real risk of unforeseen,
unpredictable problems,鈥� says Nelson Wivel, deputy director of the Institute for
Human Gene Therapy at the University of Pennsylvania, and former executive
director of the National Institutes of Health Recombinant DNA Advisory
Committee. In gene therapy, genes are ferried into cells by modified viruses or
other means. It鈥檚 a risky business, because genes can get inserted in the wrong
spot in the genome, killing the cell outright or, far worse, triggering cancer.
But at least with gene therapy there is natural damage control鈥攆ew cells
pick up the genes even when the procedure goes well, cancer only affects one
individual, and, as the procedure has always been carried out long after birth,
there鈥檚 no chance of upsetting key developmental genes.

With germ-line engineering, on the other hand, there鈥檚 more scope for
unpredictable, even monstrous, alterations. Take the so-called 鈥淏eltsville pig鈥�.
This pig, a thorn in the side of high-tech agriculturists and an icon for animal
rights activists everywhere, was engineered by scientists at the US Department
of Agriculture to produce human growth hormone that would make it grow faster
and leaner. The engineers added a genetic switch that should have turned on the
growth hormone gene only when the pig ate food laced with zinc. But the switch
failed. The extra growth hormone made the pig grow faster, but it also suffered
severe bone and joint problems and was bug-eyed to boot. Of course, unlike human
experiments, slaughtering 鈥渇ailures鈥� is always an option for animal genetic
engineers.

Before genetic engineering of humans can become a reality, each candidate
gene and its switches would need to be extensively studied in animals first, and
any changes would have to be made with a surgical precision that reduced the
chances of a 鈥淏eltsville human鈥� to just about zero. As it happens, over the past
few years, molecular geneticists have been busily developing the tools to do
just this sort of 鈥済enetic surgery鈥�.

For years, genetic engineers have altered farm animals by injecting genes
into fertilised eggs and then placing them in an animal鈥檚 womb. But the
technique is far too unreliable to use in humans. Out of every 10 000 eggs
injected, roughly three make it to adulthood with the gene functioning as
planned. What鈥檚 more, it is possible only to add whole genes, not to fine-tune
existing ones.

With mice, the process is more refined. Mice embryos contain embryonic stem
(ES) cells that will grow and divide in a flask. That allows the engineers to
make use of 鈥渉omologous recombination鈥�, the process by which DNA strands bind
to, and sometimes replace, DNA strands of similar sequence. With homologous
recombination it is possible to make tiny, surgically precise changes within
genes, with the technique depending in part on being able to sort through a
large number of ES cells, only picking out the ones that have taken the genetic
change in the correct place. Those cells are then added back to an embryo, where
they can form any part of the animal. The result is a 鈥渃himera鈥�, an animal whose
body contains both normal and altered cells. To create an animal with the
altered gene in every cell, a chimera with the change in its eggs is bred with
one that has the change in its sperm鈥攐ne reason the technique can鈥檛 be
used in humans.

But the efficiency of gene surgery is improving so that fewer cells are
needed to start with. That has made it possible for several labs to try gene
surgery directly on fertilised mouse eggs, says Dieter Gruenert, a molecular
geneticist at the University of California, San Francisco, who is developing
just such a technique. The process is still in its infancy, but it could one day
make it possible to genetically engineer human eggs, eliminating the need for
crossbreeding.

A more immediate solution will probably come from an alternative way of
generating lots of identical embryonic cells: the technology that produced Dolly
& Co.

Cloning relies on a combination of two new techniques. First, grow cells
taken from an adult or an embryo in a flask under conditions that encourage them
to divide and increase their numbers, and then trick them into reverting to a
nonspecialised state with the potential to form an entirely new individual.
Second, fuse one of these cells with an egg from which the nucleus has been
removed, and implant this cut- and-paste embryo into a womb. The wrinkles still
need ironing out, but these techniques promise engineers the luxury of an
inexhaustible supply of cells to attempt genetic surgery upon, only transferring
to an egg those nuclei they know have been properly changed. And unlike the
mouse ES cells, these cells will generate an animal with the genetic change in
every cell. Born last year, Polly, a sheep with a gene for a human clotting
factor, was created in just this manner.

Batteries of genes

With the exception, perhaps, of Richard Seed鈥攖he Chicago physicist who
in January said he would open a human cloning clinic鈥攏o one is openly
attempting to develop cloning for humans. But hundreds of genetic engineers are
working to perfect its use in other mammals, including primates. 鈥淣one of the
technologies [that will allow human engineering] is being developed only for
that purpose, but when you put them all together, that is what you will have,鈥�
says Wivel.

Even so, whether or not it is combined with cloning technology, 鈥済ene
surgery鈥� lets would-be human engineers go only so far. They could tweak a gene
here or add one there, but they couldn鈥檛 do much about characteristics such as
intelligence, say, or disease resistance, or athleticism, which are under the
control of numerous genes working in concert. For this, you鈥檒l need a budding
technology that could soon make it possible to add whole batteries of genes to
human cells.

When it comes to cell division, most of the DNA in each chromosome is
irrelevant. But to be copied properly and sorted into the two daughter cells, a
chromosome must have two types of highly specialised DNA sequences鈥攐ne
somewhere in the middle called a centromere, and bits on either tip called
telomeres. Last year, Huntington Willard, a molecular biologist at the Case
Western Reserve Medical School in Cleveland, Ohio, and his colleagues reported
that they had created artificial chromosomes in cultured human cells that
replicated every time the cells divided. 鈥淲e cultured them for six months, and
they looked like perfectly normal chromosomes,鈥� say Willard.

Because these human artificial chromosomes (HACs) promise the ultimate in
genetic engineering, they have done more to fire up discussion about human
germ-line engineering than just about any other technology. Once perfected, HACs
will make it possible for genetic engineers to ship complex custom-made genetic
programmes into human embryo cells. Each gene could come with control switches
geared to trip only in particular tissues, or when the patient takes a
particular drug.

Suppose, for instance, that men in your family tend to get prostate cancer at
a young age. Insert into your fertilised egg an HAC containing a gene for a
toxin that kills any cell that makes it, and two switches for that
gene鈥攐ne that is turned on only by prostate cells and another by ecdysone,
an insect hormone that humans cannot make. Nine months later, you鈥檙e delivered
of a bouncing baby boy. Fifty years later, he gets prostate cancer. He takes
ecdysone, which activates the prostate poison, killing every prostate cell in
his body. Even cancer cells that have spread to other parts of the body should
be wiped out.

It鈥檚 scenarios like this鈥攄reamt up by John Campbell, a molecular
biologist at the University of California, Los Angeles, who helped to organise
the March symposium鈥攖hat make the promise of human germ-line engineering
so tantalising. Hood is convinced that the benefits of germ-line engineering are
going to be substantial: 鈥淲e could probably engineer people to be totally
resistant to AIDS, or to certain kinds of cancers. We might engineer people to
live much longer. I would say all these are good qualities.鈥�

Willard agrees that the prostate cell scheme, or others like it, might
someday be made to work. At the moment, his team is trying to create HACs that
contain specific human genes so that they can check that the genes function
normally in cell cultures. 鈥淓verybody wants to [use artificial chromosomes] in
mice,鈥� he says, 鈥淏ut we鈥檙e years away from even contemplating putting HACs into
丑耻尘补苍蝉.鈥�

Still, when that day comes, as most experts predict it will, who and what
will be the first candidates for human genetic engineering? Geneticists are more
willing to kick around the possibilities than ever before. Gruenert speaks for
many when he says that the crucial issue is whether germline engineering would
save lives and prevent suffering. 鈥淔or medical reasons, I have no problems,鈥�
says Gruenert. 鈥淏ut for making superwomen or supermen? I have some problems with
迟丑补迟.鈥�

The first candidates for human genetic engineering are likely to be children
who could inherit a disorder that kills young, is incurable both now and for the
foreseeable future, and is caused by a relatively simple defect. Tay-Sachs
disease, which causes the brain to degenerate in the first few years of life, is
just such a disease. Fix the gene, goes the argument, and you stop the disease
both in that child and all his or her offspring.

If the safety issues are resolved, the idea of wiping out such diseases could
sway the opinions of the public and regulatory agencies, paving the way for the
first attempt at human germ-line engineering, says Jeremy Rifkin of the
Foundation on Economic Trends in Washington DC, a longstanding opponent of
biotechnology. 鈥淚鈥檝e seen this pattern before in biotech. First there is some
discussion in journals, then a conference, then they go ahead and do it. I think
there are protocols being readied now, and we鈥檒l see them within a year or
迟飞辞.鈥�

Strange bedfellows

Rifkin may have some unusual allies in his fight against human engineering.
Many researchers who are otherwise decidedly pro-biotechnology are vocal in
their concerns about engineering humans. Allen Roses, who heads Glaxo Wellcome鈥檚
worldwide genetics research effort, is emphatic that any attempts at germ-line
engineering would be 鈥渕orally questionable鈥�. The milder-mannered Francis
Collins, director of the National Human Genome Research Institute near
Washington DC, says simply: 鈥淚t is very hard to come up with compelling
scenarios of why you鈥檇 want to.鈥�

Collins, Roses and others take issue even with the idea of using genetic
engineering to prevent genetic disorders. They point out that parents known to
be at risk of certain serious genetic abnormalities are already offered genetic
testing and the option of an abortion if their fetuses have the disorders. Using
this approach, the number of Tay-Sachs births has been reduced by more than 95
per cent among American Jews. For women willing to have IVF, an embryo can even
be tested before pregnancy starts.

Of course, repairing rare genetic defects is not the only factor likely to
endear genetic engineering to the public. Who could resist the chance to
bequeath their children freedom from Alzheimer鈥檚, cancer, heart disease and
diabetes?

Then there鈥檚 the possibility of cosmetic changes and enhancements
that have nothing to do with saving lives and preventing disease. Many
behavioural traits, from cheerfulness to sexual orientation, have already been
linked, if tenuously, to variations in single genes. Many more such links will
be reported in the near future. 鈥淭here will come a time when we will understand
enough to manipulate even complex genetic systems,鈥� says Hood. 鈥淔or example, we
will be able to dramatically affect intelligence. That, I think, will be pretty
颈谤谤别蝉颈蝉迟颈产濒别.鈥�

鈥淓volution is being superseded by technology, and the time scale will be far
more rapid,鈥� says Stock. 鈥淗umans are becoming the objects of conscious
诲别蝉颈驳苍.鈥�

And no matter how wild the idea of designer children sounds now, technology
has a way of making believers out of sceptics. Silver argues that parents will
provide the market forces that will eventually make germ-line engineering of
humans routine. When IVF was first being developed in the 1970s, he points out,
doctors and lay people alike thought the idea absurd and repellent. Even though
success rates are still low, IVF created such demand among couples with
fertility problems that it has become widely accepted and commonplace.

For now, however, the regulative barriers are firmly down, even as the
research hurtles forward with breathtaking speed. Which is perhaps why talk
about engineering humans is now coming into the open. It no longer makes sense
to shy away from discussing what we鈥檙e going to do when all the technical
obstacles are overcome, and genetic engineering offers us the profound power to
sculpt our children鈥攁nd the future of our species.