快猫短视频

DEEP in the brains of some unfortunate British cows, changes are afoot. As
normal proteins transform into sticky clumps, healthy calves become mad cows.
Then dead ones. And although the strange 鈥減rion鈥� proteins that are responsible
have been cast as biological deviants because their only known function was to
wreak neurological havoc, under the weight of new evidence that thinking is
being transformed, too. Prions, it turns out, have a role to play, perhaps a big
one, in healthy creatures as well as diseased ones.

The sea change in attitudes towards prions began when Reed Wickner, a
geneticist with the National Institutes of Health near Washington DC, discovered
prions in yeast. Rather than causing disease, Wickner鈥檚 prions were passing on
genetic traits from one yeast generation to the next, a job which for 50 years
has been considered the exclusive province of DNA.

The story might have ended right there鈥攁n astounding anomaly, to be
relegated to a textbook footnote鈥攅xcept that there have been more
sightings of these weird prion genes. Besides giving geneticists pause for
thought, the new revelations are forcing biochemists to rethink their ideas
about how proteins behave, and may compel developmental biologists to entertain
new ideas about how embryo growth is regulated. And although prion genes have so
far only put in an appearance in fungi, some say they will soon be popping up
all over, even in humans.

鈥淓very organism will have them,鈥� predicts Susan Liebman, a geneticist who
studies yeast at the University of Illinois. 鈥淭hey will not be rare.鈥�

Curly-haired yeast

A heritable trait in the yeast Saccharomyces cerevisiae, whether
passed on by DNA or anything else, is not as easy to spot as curly hair, big
feet or blue eyes. Under the microscope, a yeast is a yeast is a blob. Yeasts
do, however, metabolise chemicals in different ways. Most can鈥檛 get their
nitrogen from a chemical called ureidosuccinate whenever ammonia is available.
But some yeasts, called [URE3] and pronounced 鈥測uri-3鈥�, happily digest
ureidosuccinate in the presence of ammonia. The [URE3] types of yeast show up in
any population at the very low rate of about one in 100 000, and the way they
pass on their ureidosuccinate-gobbling trait totally disobeys the most
fundamental rules of inheritance.

Most of the time, yeasts reproduce by budding. But when the moment is right,
they have sex. When yeast cells join, the union is total. The cells fuse, mixing
everything: cytoplasm, proteins, nutrients, genes and cell walls. The fused
parental cell then divides into four new daughter cells, distributing most of
its component parts evenly鈥攅xcept for the genes. Each parent contains one
each of the 16 yeast chromosomes, and each offspring gets one copy, chosen from
the parental mix like a ticket from a raffle tumbler. That way if one parent has
a trait that the other lacks, only half the offspring get it. It鈥檚 a simple
rule, but one that [URE3] seems to ignore.

Rather than appearing in just half of the progeny of a yeast coupling in
which only one of the parents has the trait, [URE3] appears in all four
daughters. Yeasts have a handful of other traits that disobey Mendel鈥檚 laws, but
these are carried on the genes of parasitic viruses or bacterial plasmids that
infect all progeny when a yeast cell divides. No viruses or plasmids have ever
been found that could account for [URE3]鈥檚 strange pattern of inheritance, or
that of a second trait called [PSI], in which the cells make proteins that are
too long. By 1989, when Wickner began working on [URE3], most researchers had
given up looking for explanations to such riddles. 鈥淚 felt like the den mother
of these non-Mendelian elements, just because nobody else cared,鈥� says
Wickner.

At that time on the other side of the Atlantic, the epidemic of 鈥渕ad cow
disease鈥� or bovine spongiform encephalopathy was raging as furiously as the
debate about its cause. At the core of the debate was a proposal from Stanley
Prusiner of the University of California in San Francisco suggesting that
proteins called prions caused a number of related brain diseases, including
scrapie in sheep, mad cow disease, and, in humans, Creutzfeldt-Jakob disease.
Prion proteins can flip from their regular shape to the disease-causing shape,
and鈥攊n a domino effect鈥攆orce proteins of the same composition to
switch as well. The distorted proteins form insoluble clumps that gum up and
eventually kill neurons, opening cavities in the brains of afflicted
creatures.

The most controversial part of Prusiner鈥檚 theory is that the prions are
infectious, passing from one animal to the next, carrying their
protein-distorting powers with them. Prusiner won one of last year鈥檚 Nobel
prizes for his work, but there are still those who argue that the 鈥減rotein only鈥�
theory of infection is plain wrong. They say that infections can only be carried
by DNA (or its RNA counterpart) contained in microorganisms like bacteria and
viruses.

Wickner was inclined to believe Prusiner. He, too, was beginning to doubt
DNA鈥檚 omnipotence, and if prions worked as Prusiner said they did, they could
explain how all the offspring of a yeast coupling can inherit a trait that only
one parent had. But it was just a hunch, so he didn鈥檛 involve any of his
postdocs. 鈥淲hy risk a young career? It was a long shot,鈥� says Wickner.

His idea went like this. A single distorted version of a prion protein could
rapidly convert all the other proteins of the same composition in a yeast cell.
When that cell fused with another, these prions in turn would sweep across the
fused parental cell, distorting all the normal copies of the protein, and
passing them on to the four daughter cells and every other descendant of the
original cell down the generations. Wickner鈥檚 candidate for such a protein was
one already known to researchers called Ure2.

Yeast cells lacking the gene for the Ure2 protein behave just like [URE3]
yeast, except that they pass on the trait in proper Mendelian fashion to just
two progeny, suggesting that a malfunctioning Ure2 protein could be at the heart
of the ureidosuccinate-gobbling trait in [URE3].

To test his hypothesis, Wickner crippled the Ure2 gene in [URE3]
yeast and let the cells grow for a few generations. If he was right, any
inherited molecules of the distorted, presumably malfunctioning, form of Ure2
would break down as all proteins do in time. With no new Ure2 being made, the
prions and the [URE3] trait would vanish.

To test whether he had successfully eliminated the distorted protein from the
yeast colony, he reinserted the Ure2 gene. As he had predicted, the
Ure2 protein started working normally again, and the whole population of yeast
was cured of [URE3]. Just as a cow must be infected with a prion 鈥渟eed鈥� to go
mad, yeast needs a prion seed to create the [URE3] trait.

Instant sensation

The more tests Wickner did, the more it looked as if his weird genetic trait
was carried by prions. Add guanidinium chloride to brain extracts from sheep
with scrapie and the prion clumps break up, with the extract rendered
noninfectious. Similarly, when Wickner added guanidinium chloride to his brews
of [URE3] yeast, they were cured of the trait. What鈥檚 more, says Wickner, [URE3]
spontaneously reappeared in a small number of cured yeasts. A virus would have
been gone for good.

When Wickner announced in Science in 1994 (vol 264, p 566) that
yeast can use prions instead of genes to pass on heritable traits it caused an
instant sensation. 鈥淣othing I鈥檝e ever done in my life has brought forth such
enthusiasm,鈥� says Wickner. One commentator proclaimed that the finding gave
mammalian prions 鈥渁dditional respectability鈥�. Even researchers like pathologist
Laura Manuelidis of Yale University, who are critical of the notion that prions
are infectious in mammals, said that they found the story convincing.

Fellow yeast geneticist Susan Lindquist at the University of Chicago set out
to confirm Wickner鈥檚 findings. And a little over a year ago, she reported that
just like [URE3] yeast, [PSI] yeast cells are cured of their trait by
temporarily stopping a gene, in this case for a protein called Sup35. She also
tagged the Sup35 protein so it glowed green and showed that under the microscope
the luminescent protein formed clumps in [PSI] yeast. As expected, when those
yeasts mated, they passed on the glowing green clumps to all four offspring.

But it was Liebman and Michael Ter-Avanesyan at the Cardiology Research
Centre in Moscow who found evidence that prions are part of being a normal,
healthy yeast, rather than a sign of some yeast equivalent of mad cow
disease.

The Sup35 protein comes in two parts. Most of the Sup35 protein is needed for
its regular job of halting the cell鈥檚 synthesis of new proteins when they reach
the proper length, but a small portion at the head of the Sup35 protein is
entirely dispensable for that function. Yeasts lacking this extra bit are
perfectly healthy, but, Ter-Avanesyan found, they can never be converted to
[PSI]. He concluded that the dispensable portion was necessary for the
non-Mendelian trait.

Liebman鈥檚 lab confirmed the findings of Ter-Avanesyan and went one step
further. For reasons that are still poorly understood, a blast of full-length
normal Sup35 protein converts normal yeast to [PSI] yeast. Liebman found that a
blast of the Sup35 small domain alone is sufficient to convert yeast into
[PSI].

It turns out that the Ure2 protein also has a portion whose sole job is to
enable the protein to switch between its two forms. Clearly, yeasts aren鈥檛
producing prions by accident: dedicated parts of their proteins and genes make
it possible for the proteins to flip from the soluble form to the clumping
form.

Flip-flop

Last May, Lindquist published the first electron microscope photos ever of
yeast prions, in this case Sup35, in their insoluble aggregates. The long,
fibrous clumps looked almost identical to those seen in the brains of cows with
mad cow disease and humans with Creutzfeldt-Jakob disease, adding even more
weight to the idea that, far from being confined to the brains of sick mammals,
prions happen in healthy yeast, and very likely other species, too.

鈥淚f you have something in yeast, and you have it in humans, you probably have
it in everything,鈥� says Liebman. Peter Lansbury, a biochemist at the Harvard
Medical School in Boston who studies mammalian prion diseases, agrees. Once
people start searching for prions in other species, they鈥檒l find them, he
predicts鈥攁s long as they know what to look for, that is.

One problem critics have always had with the idea of prions, whether they be
passing on heritable traits or causing fatal brain diseases, is that proteins
are supposed to be stiff, not flexible enough to flip from one state to another.
鈥淏ecause the first proteins described were very rigid, people have this idea
that proteins are like rocks,鈥� Lansbury says. 鈥淏ut it鈥檚 very clear now that
proteins are flexible.鈥� So if you are hot on the trail of a prion in some
species other than a yeast or a mad cow, don鈥檛 ignore the fact that a prion
protein will exist in more than one form.

To make the hunt slightly more challenging, prions, in their distorted form,
tend to clump together. These clumps don鈥檛 dissolve in water, so the prions get
left behind, stuck to the bottom of the test tube, while the soluble form of the
protein gets studied. 鈥淧eople spin down their samples,鈥� says Lansbury. 鈥淚t鈥檚
common practice. If it aggregates it鈥檚 hard to study, so you throw it away.鈥�

Of course, the effort required to track down prions raises the issue of
whether it鈥檚 all worth it. What interesting things could the prions possibly be
doing if they do turn out to exist in a wide range of living organisms? Prions
are not independent entities like viruses or bacteria engaging in their own
struggles to survive, but parts of living creatures. If their only role was to
cause disease, evolution would have got rid of them. Lindquist suspects that
[PSI] and [URE3] prions help whole yeast populations respond rapidly to changes
in their environment. The [PSI] trait, for example, is more likely to occur
spontaneously when a population is stressed, say by high temperature, suggesting
that the trait may have something to do with helping yeast survive hard times.
To check out that idea, Lindquist is now searching for genes that only function
in [PSI] yeast.

Fred Cohen, who works with Prusiner at the University of California in San
Francisco, also thinks that evolution will have found a way to make good use of
prions, not least because of early hints that prions are far more common than
anyone suspected. In a computer analysis of more than 100 000 theoretical
proteins, Cohen and Prusiner have found that up to 3 per cent have two stable
conformations, one of which can clump in true prion fashion. 鈥淚 can鈥檛 think of
any detrimental phenomenon in biology that can鈥檛 somehow be used to advantage in
another setting,鈥� says Cohen.

The first case of prions being positively demonstrated to help an organism
survive came in September, from French researchers working with a fungus called
Podospora anserina (Proceedings of the National Academy of
Sciences, vol 94, p 9773). The fungus grows in spreading mats on forest
detritus. When one fungal mat encounters another, the cells at the edges fuse,
much the way mating yeast cells do, mingling their contents. The trouble is that
if one colony has a virus, both soon get infected. To minimise the damage,
sometimes the fused cells sacrifice themselves for the good of the colony, dying
and forming a barrier that stops the virus from spreading.

What determines whether the barrier forms is the prion-like transformations
of a protein called Het-s, says Joel Begueret, a biologist at the Institute of
Biochemistry and Cellular Genetics in Bordeaux. When a colony containing one
form of the Het-s protein fuses with a colony containing the other, all the
fungi quickly transform to a single form, as if a prion infection has swept
through the colonies. The fused colony can then form the dead-cell blockage when
it fuses with another colony.

Fine for fungi

Begueret has ruled out the possibility that a virus is responsible or that
one colony is switching on genes in the other colony. What鈥檚 more, if you add an
artificial gene to the fungus to produce a blast of the normal form of the Het-s
protein, the fungus converts to the form that makes the dead-cell blockage. 鈥淚t
sounds very prion-like,鈥� says Cohen.

Fine for fungi, but what about complex organisms such as humans? Do they used
prions to pass on useful traits? No one is yet prepared to say that prions may
be carrying heritable traits from human parents to their offspring, although
theoretically it is possible. But prions may be involved in the elusive
mechanism called imprinting that allows genes to behave differently depending on
whether they are inherited from your mother or your father, says Lindquist (see
鈥淲here did you get your brains?鈥� 快猫短视频, 3 May 1997, p 34).

Developmental biologists like Igor Dawid at the National Institutes of Health
near Washington DC are happy to toy with the notion that prions may help the
fertilised egg, an amorphous single-celled blob, become a multicellular organism
with wings, livers, arms or antennas.

At some point in this transformation, the embryo cells decide whether to
become liver, muscle or any other tissue. Once the decision is made, all their
progeny have to stick with it. Yet all cells carry the same genetic code, so
it鈥檚 unclear how cells are passing on that information.

Prions could be responsible. Cells in different tissues manufacture different
types of proteins, some of which keep the correct genes for that tissue turned
on. If those regulatory proteins had the ability to spread their influence as
prions do, then dividing cells within a tissue would automatically know which
genes to turn on. 鈥淣o one would say that all such phenomena have to do with
prions,鈥� says Dawid. 鈥淏ut some of them could, and that would be very
别虫肠颈迟颈苍驳.鈥�

Certainly, when a group of cells in a developing animal or plant embryo sets
off to become a particular type of tissue, large amounts of the regulatory
proteins appear in a burst that later subsides. As fungi show, a burst of normal
protein is often all it takes to trigger the appearance of its prion forms.
Lindquist has even identified regulatory proteins that contain sequences
resembling the bits of the yeast prions that allow them to switch between their
two shapes, although she says its still too early to give specifics.

Meanwhile, if Lindquist, Lansbury and Liebman are right about the
wide-ranging ramifications of the new 鈥減rotein-only鈥� genetics, throwing out
uncooperative protein clumps may no longer be a wise option. Everyone from
developmental biologists to yeast geneticists should probably start paying
attention to the gunk stuck at the bottom of the tube.