午夜福利1000集合

SUPPOSE you could travel back in time and see inside your mother鈥檚 womb as
the fetal you sprouts fingers and toes. If you looked closely, you would find
something alarming.

Your tadpole body, which you would expect to be insulated from the filthy
world outside, is contaminated with nasty-looking viruses not unlike the AIDS
virus. Meanwhile, billions of your cells are diligently pumping out yet more of
these parasites, as if in mutiny. Peer inside your placenta and things look even
worse. It鈥檚 positively teeming with viruses, dumping legions of the tiny
invaders into its surroundings.

Should you be amazed that you ever survived such an infection? Surprisingly,
any embryologist would reassure you that it wasn鈥檛 an infection at all, but
rather a normal part of every pregnancy. These HIV lookalikes, dubbed endogenous
retroviruses or ERVs, are actually encoded in the DNA of every mammal. They
invaded cells during infections millions of years ago and liked it so much they
decided to stay. Most surprising of all, some researchers now argue that ERVs
may have played a starring role in the evolution of mammalian life and its
crowning achievement鈥攍ive birth. ERVs, they say, were critical to the
emergence of both the placenta and the mechanisms that protect the fetus from
pathogens and the mother鈥檚 immune system. And although the researchers don鈥檛
exactly agree on the details, without ERVs, they say, we humans might still be
laying eggs rather than just eating them.

The advantages of gestating your offspring inside your body rather than in an
egg may not be readily apparent until you consider the staggering success of
mammals. They occupy every ecological niche on land and in the sea, from the
poles to the tropics. One group, the bats, has even taken to the air. While
marsupials give birth to undeveloped young that mature in the pouch, and two
mammals鈥攖he duck-billed platypus and the spiny anteater鈥攃ontinue to
lay eggs, the vast majority give birth to well-developed live offspring. That
ability, some biologists argue, is one of the two key factors that gave mammals
the edge over birds, reptiles and fish when it came to competing for ecological
niches vacated by the demise of the dinosaurs (the other was having a
fast-moving, warm-blooded metabolism). Developing inside the body rather than
inside an egg means that the fetus can eliminate waste and receive the
continuous, generous supply of oxygen and nutrients that is needed to build such
energy-hungry features as a large mammalian brain.

Though we take live birth for granted, before you can think of growing a
fetus to an advanced stage of development inside its mother鈥檚 body, a few
potential problems need sorting out. The most obvious is immune rejection. Half
the fetus鈥檚 genes come from the father, meaning the mother鈥檚 immune system
should recognise it as foreign and chew it up. As if that weren鈥檛 provocation
enough, the placenta grows like a tumour, invading the uterus wall and even
sending nodules of genetically foreign cells to grow in distant parts of the
mother鈥檚 body. Despite that goading, the fetus manages to coexist with its
would-be attackers.

Intimate contact

It must also protect itself against any viruses and other pathogens that
infect the mother, because they can have extreme consequences for a fetus. If
the infection hits during the development of the nervous system, for instance,
the baby may be born deaf or mentally handicapped.

This creates a Catch-22 situation for the mammalian fetus. To defend itself
against rampaging immune cells and pathogens, it could build a firewall between
itself and the mother鈥攖he more impermeable, the better. Yet the fetus must
also maintain intimate contact with its mother in order to enjoy the benefits of
nutrition, oxygen and waste disposal. Most reptiles, fish and amphibians that
give birth to live young get around these problems by forgoing the advantages of
nutrient and waste exchange and protecting the fetus with a fairly impermeable
barrier. Mammals have come up with a better solution. They have evolved a
placenta which separates the blood supplies of the mother and fetus by just a
few layers of cells, usually including a 鈥渟yncytium鈥� layer in which cells are
fused together into a vanishingly thin sheet. The thinness of the barrier
ensures that nutrients, oxygen and waste can diffuse across, while the fusion of
the placental cells鈥攖ogether with some other more subtle
embellishments鈥攅nsures that most immune cells and pathogens cannot worm
their way into the fetus.

This, so the argument goes, is where the ERVs come in. When live birth began
to evolve, some 120 million years ago, the ancestor of all live-bearing mammals
made use of one of its ERV parasites to evolve a highly sophisticated
placenta.

Placental ERVs were first discovered in baboons in the early 1970s. The
researchers were using electron microscopy to work out whether the rubella virus
could cross into the unborn animal. But instead of rubella, they were shocked to
find tiny, spheroid retroviruses budding from the cell surfaces of the placenta,
even in apparently healthy animals. Retroviruses, like all other viruses, can鈥檛
replicate under their own steam. Instead, they insert their genes into the DNA
of the cells they infect and then instruct them to manufacture viral particles,
which bud from the surface of the cell en route to infecting their next victim.
Because ERVs insert their genetic blueprints into the cells that produce sperm
and eggs, they are passed effortlessly from one generation to the next without
ever having to infect another host.

Virologists have spotted ERVs in the genome of every mammal they鈥檝e checked,
and we humans are known to harbour at least a thousand ERVs, many of which have
hitchhiked with us for well over 30 million years. ERVs spend most of their time
asleep, present only as extra segments inserted here and there in our DNA. But
in the placenta and in some fetal tissues, a select handful awaken, commanding
the cells that harbour them to produce ERV proteins and assemble them into
retroviruses.

Once a baby is born, ERVs fall pretty much silent. Occasionally, however,
they rear their heads in human diseases: they may allow certain tumours to grow
more aggressively or play a role in autoimmune diseases such as lupus. This
raises the question of why these stowaways have been retained through mammalian
evolution. Perhaps ERVs also play some beneficial role鈥攐ne important
enough to offset the evolutionary price of keeping them around (see 鈥淲here鈥檚 the
辫补测辞蹿蹿?鈥�).

Erik Larsson, a pathologist at the University of Uppsala in Sweden, suggested
in 1988 that they are involved in protecting the fetus in two ways: by helping
the placental cells to fuse into a syncytium and by suppressing the immune cells
in the immediate vicinity. He based his theory on similarities between the
envelope protein that studs the outer surface of HIV and other viruses, and ERV
versions of this protein.

When HIV-infected cells produce the envelope protein, they can fuse with
uninfected cells, helping to spread the infection. In much the same way, Larsson
suggested, producing an ERV鈥檚 envelope protein helps cells of the placenta,
called trophoblasts, to fuse to form the microscopically thin syncytium.

That idea sounds especially attractive when you consider that in humans it is
relatively rare for cells to fuse to form what amounts to a single giant cell.
Sperm and eggs, for example, fuse during fertilisation, muscle cells fuse to
form skeletal muscle, and monocytes鈥攁 type of immune cell鈥攆use to
form huge cells called osteoclasts that devour damaged bone. Interestingly,
monocytes also produce ERV envelope protein.

But the best evidence for Larsson鈥檚 hunch comes from studies on placental
cells that can be grown in the lab called BeWo cells. BeWo cells can be
chemically induced to fuse into a syncytium similar to that seen in many
placentas. When they do this, they produce bucketfuls of the envelope protein of
the best-known human placental ERV, called ERV3. Conversely, preventing the BeWo
cells from producing ERV3 envelope inhibits the formation of a syncytium, while
treating them with a segment of DNA that triggers the production of ERV3
envelope will make them undergo some fusion in the absence of any chemical
prod.

Pushing the envelope

Neal Rote, a reproductive immunologist at Wright State University in Dayton,
Ohio, who conducted these studies (Placenta, vol 20, p 109), believes
that ERV proteins play an even larger role than simply getting cells to fuse.
His team has also found that triggering production of ERV3 envelope persuades
the BeWo cells to take on many of the characteristics of syncytium cells, such
as an enlarged nucleus. 鈥淚t surprised us quite a bit,鈥� says Rote, 鈥渢o see that
ERV3 [envelope] actually controls other aspects of differentiation. We never
expected that.鈥�

Besides helping the placenta to form in the first place, ERV envelope may
help protect the fetus from rejection by locally suppressing the mother鈥檚 immune
system. In support of that theory, Joachim Denner, a molecular biologist at the
Paul Ehrlich Institute in Langen, Germany, has found that the envelope protein
from one ERV, called HERV-K, is a potent suppressor of immune cells. Incubating
immune cells like B cells and T cells with parts of the envelope protein
inhibits chemically induced cell proliferation. The same envelope protein
segments also suppress the cells鈥� production of interleukin-2, which can perk up
the immune system, and increases their production of interleukin-10, which tends
to calm the immune system.

Thierry Heidmann, a molecular biologist with the Gustave Roussy Institute in
Villejuif, France, has even more striking evidence for the ability of envelope
proteins to suppress the immune system. When he injected tumour cells into
normal mice, their immune system rejected them. However, when the tumour cells
were genetically modified to produce the envelope protein of a mouse retrovirus
they escaped immune rejection and grew more aggressively (Proceedings of the
National Academy of Sciences, vol 95, p 14 920). Certain tumours of the
ovaries and testes produce ERV particles, Heidmann points out, which suggests
that they may actually harness ERVs to evade destruction by the immune
system.

For his own part, Larsson says his theory explains why only one of the three
genes that make up ERV3鈥攖he envelope protein gene鈥攈as survived
intact in the human genome. Mutations have rendered the other two genes
incapable of making proteins. The envelope gene may have survived evolution
because it is the only part of ERV3 that its host needs鈥攆or its
contribution to placental function. 鈥淵ou don鈥檛 need a replicating virus, you
just need one protein,鈥� says Larsson. Otherwise, selective conservation of
envelope protein makes no sense, he says.

Still, even proponents of Larsson鈥檚 theory acknowledge that the evidence is
largely circumstantial. What鈥檚 more, they say, there is another explanation for
why placental and fetal tissues produce large quantities of ERVs: it could be a
throwback to the time when the virus needed to get an infection going just as
the fetus鈥檚 sperm or eggs were starting to develop, to ensure it was passed to
the next generation. 鈥淔rom the virus鈥檚 perspective,鈥� says Patrick Venables, a
rheumatologist at the Kennedy Institute of Rheumatology in London, 鈥減lacental
and fetal [gene] expression happens because evolutionarily, if you want to get
inserted into the germ line, then you鈥檝e got to be active during gestation.鈥�

To some, the data showing that the ERV3 envelope gene has been preserved
during human evolution are also suspect. That the gene remains grossly intact
and so capable of producing a protein of some sort is obvious because no major
mutations have disrupted its coding sequence. But as Heidmann points out, minor,
difficult-to-detect mutations which exchange one amino acid for another may have
eroded the protein鈥檚 ability to function properly in a live animal. 鈥淚t鈥檚 not so
easy,鈥� he says, 鈥渢o say ERV3 [envelope] is really functionally conserved.鈥�

Potentially even more damaging to the theory that the envelope protein is
essential for placental growth is Heidmann鈥檚 recent discovery that 1 per cent of
white people harbour a defective version of ERV3 that produces a severely
shortened envelope protein (Journal of Virology, vol 72, p 3442). The
shortened protein lacks the parts responsible for cell fusion and immune
suppression. Nonetheless, individuals carrying only the mutant ERV3 are born
healthy and reproduce normally.

Essential backup

Larsson, however, has a comeback. He points out (and most of his critics
agree) that the case for ERVs being essential for placental function does not
rest on ERV3 alone; those studies merely illustrate a point. Evolution, he says,
would protect crucial functions with built-in redundancy, so if one gene is
lost, another can fill in. This could be true for ERV3. Although most of the
thousand or so other ERVs present in human DNA are defective, a few鈥攕uch
as HERV-K鈥攑ossess intact envelope genes. No one knows whether or not women
with a shortened envelope gene for ERV3 still spew out virions during pregnancy,
so this hunch has yet to be tested.

As for immune suppression, the fetus has several other ways to protect
itself. For example, it produces proteins such as HLA-G and fas ligand, which
deactivate certain immune cells. 鈥淚 think there鈥檚 going to be a hundred
different mechanisms of how immune suppression occurs during pregnancy,鈥�
Venables says.

Even Heidmann, while insisting that his new finding disproves the idea that
ERV3 envelope protein is essential for placental functions such as cell fusion
and immune suppression, is happy to concede that the protein may perform other
roles, such as protecting the fetus from new retroviral infections. ERV proteins
could stop HIV and other retroviruses from entering the cells of the fetus by
binding to and monopolising the cell-surface receptors which the viruses need to
get in. This is known as receptor interference. Such a mechanism could explain
why, provided efforts are made to prevent a baby coming into direct contact with
its mother鈥檚 blood during birth, HIV is relatively unlikely to be transmitted
from an infected mother to her baby.

So the debate rattles on, and placental biologists continue to puzzle over
just how mammals evolved that mysterious throwaway organ that enables them to
cart their developing young around, while pumping them full off nutrients and
oxygen. True, no one is prepared to bet the farm on the ERV theory, but for the
moment at least there is a dearth of alternative explanations for how the
mammalian placenta came to be. And, of course, the ERV explanation has comfort
value too鈥攊t helps to explain why the placenta of a healthy, blossoming
pregnant woman is riddled with those mysterious viruses.

The genomes of humans and other vertebrates are riddled with the relics of
retroviruses that first infected them millions of years ago. These endogenous
retroviruses (ERVs) may provide some services for their hosts (see 鈥淲here鈥檚 the
payoff?鈥�), or they may once have helped their host to gain resistance
to infection by exogenous strains of the same or related retroviruses. But there
could be a more insidious side to ERVs.

Once an ERV has infected the genome of a sperm or egg it is automatically
passed on from generation to generation. Freed from the need to replicate, the
ERV genes accumulate random mutations. What鈥檚 left in the host鈥檚 DNA looks like
a junkyard of used retrovirus parts鈥攁nd that, according to some
biologists, could spell trouble.

Animals regularly encounter new retroviruses, and when an infection occurs
the retrovirus may swap genetic parts with defective ERVs. 鈥淵ou generate a whole
lot of new viruses with different properties,鈥� says Leonard Evans at the
National Institute of Allergy and Infectious Diseases in Hamilton, Montana.
鈥淸They] infect different cell types and use different receptors.鈥�

Evans is not merely speculating. In mice, the Moloney murine leukaemia virus
induces leukaemia through a multistep process in which it produces a new hybrid
virus by swapping parts with the mouse鈥檚 ERVs. Another group has found that in
one strain of mice, a defective ERV regains its ability to replicate by swapping
parts with another defective ERV. The newly formed virus may be responsible for
the high rate of lymphoma in these mice.

No one knows for sure whether human retroviruses have ever permanently
swapped genetic parts with ERVs to create new viruses. However, there are
suspicions that an ERV protease enzyme might temporarily substitute for the HIV
protease when it is under attack from AIDS drugs called protease inhibitors
(This Week, 19 December 1998, p 21).

Endogenous retroviruses (ERVs) are pathogens whose genetic blueprints were
trapped in the DNA of our ancestors during infections millions of years ago.
Most ERVs are now silently carried in the genome. But in some parts of the body
a few ERVs continue to produce proteins鈥攁nd even whole viral particles.
That has led to speculation that ERVs have been commandeered by the animals they
infected to carry out certain tasks, in some ways similar to the bacteria that a
billion years ago sought refuge inside larger host cells and became mitochondria
and plant chloroplasts. 鈥淐ells don鈥檛 like to waste a lot of energy making
proteins unless they have a function,鈥� says Neal Rote of Wright State University
in Dayton, Ohio.

One line of evidence suggests that ERVs help the placenta to form a barrier
between mother and fetus (see main story). The immune system may also use ERVs
to regulate itself. Immune cells called monocytes shed ERV proteins, which in
lab tests have been shown to alter the activity of other immune cells.

Finally, ERV proteins spew from the surface of cells that line the gateways
to the body such as the lungs and placenta, as well as from the sebaceous glands
that secrete oils onto the skin. One possibility is that virions help to block
infections by competing for the receptor sites that pathogenic viruses use to
enter cells.

For now, evidence that ERVs play an active role in the biology of mammals is
still speculative. 鈥淏ut modern biology accepts the idea of endosymbiosis with
mitochondria,鈥� says Timothy Lyden of Ohio State University in Columbus. 鈥� I
don鈥檛 see why we would expect ERVs to be different.鈥�