Here is the article from the Economist

[Guest guest]

Straining belief

WHAT is the difference between " protein " and

" protean " ? Nothing, according to Stanley

Prusiner, if the protein in question is a prion. Dr

Prusiner, who works at the University of California, San

Francisco, has championed the idea that a range of brain

disorders*including Creutzfeldt-Jakob disease (

CJD ) in man, bovine spongiform encephalopathy

( BSE , or " mad-cow disease " ) in cattle, and

scrapie in sheep*are caused by misshapen

proteins called prions. These, the theory goes,

transmit disease from one organism to another in the

same way that viruses do. The prion hypothesis defied

biological orthodoxy when Dr Prusiner advanced it in

1982, because no genetic material ( DNA or RNA ) was

involved in the infection. But he has managed so

thoroughly to persuade the scientific powers-that-be of

its truth that he won last year's Nobel prize for

medicine. Now, his research team has found evidence

that helps to fill one of the remaining holes in the

hypothesis*how a single species can suffer from

several apparently different sorts of prion disease. At the

same time, he has added to the armoury of tests

available to those who study such diseases.

A twist in the tale

The protein from which prions are formed (known,

with startling originality, as prion protein) is found in

almost all tissues in man and

beast. In its " normal " form,

it goes about important but

still ill-defined business in

cells. When bent out of

shape, however, it can cause

disease.

Shape is crucial to the

function of all types of

protein, but the body is

normally able to tolerate a

few abnormal protein

molecules. What makes

prion protein unique,

according to the Prusiner

model, is that misshapen

molecules of it somehow

cause well-formed ones to

become misshapen too.

That creates a

chain-reaction that deforms

much of the prion protein in

an individual. These

misshapen prions

accumulate in the brain (as

well as in some other

tissues, including spleen and

possibly blood). This, in

turn, causes widespread

damage to the brain*and,

ultimately, death. It also

means that if a malformed

prion passes from one

individual to another, the

disease can spread in the

same way as a more

traditional infection. Indeed,

prions are so infectious that

the recipient need not even

be of the same species for

the disease to take hold.

This is all plausible and is

now supported by much

experimental evidence. But

it still does not explain how,

given that there is only one

sort of prion protein, several

different types of prion

disease can afflict a single

species. There are, for

example, at least eight

strains of scrapie,

distinguishable by their

incubation periods and by

the different areas of the

brain that they damage. And

in people, the " new variant "

of CJD (known as nv CJD ),

which all available evidence

suggests is a human version

of BSE , is palpably distinct

from the disease that

Creutzfeldt and Jakob first

described.

The idea most consistent

with Dr Prusiner's original

hypothesis is that more than

one malformation of the

prion protein is possible, and

that the process by which a

malformed protein subverts

a healthy one is so finely

tuned that each

malformation is copied

faithfully along the chain of

infection. The slight

differences in shape of the

resulting prions would alter

their biochemistry enough to

produce the distinct sets of

symptoms that accompany

different strains.

An extraordinary hypothesis,

you might think. But a paper

just published in Nature

Medicine by a colleague of

Dr Prusiner, Jiri Safar, and

his team adds yet more

weight to the evidence that it

is true.

Dr Safar's shape-detector is

an antibody. Antibodies are

proteins that are exquisitely

sensitive to shape (their role

is to stick on to, and thus

neutralise, the " foreign "

molecules in viruses,

bacteria and the like, and

they do this by recognising

the foreigners' shapes).

Even slight changes in the

shape of a protein may

affect how well a particular

antibody sticks to it. Dr

Safar's antibody attaches

itself strongly to normal,

healthy prion protein, but

seems to bind with variable

strength to different

disease-causing prions. The

differences are enough to

distinguish eight strains of

prion in hamsters*a

favourite experimental

animal for prion researchers.

Dr Safar and his team now

hope to extend the test to

prions in other species, and

specifically to the one that

causes BSE and nv CJD . At

the moment there are two

strain-sensitive tests for this

prion. One, perfected by

Moira Bruce and her

colleagues at the Institute for

Animal Health in Edinburgh,

depends on infecting mice

with prions and observing

what happens. But a single

test involves many mice and

many months of hard work,

and the process costs

30,000 ($50,000) per cow

(or human) tested.

The second test, developed

by Collinge and his

team at Imperial College,

London, is done in a test

tube. It relies on the ability

of a particular enzyme to

chop prions into

different-sized bits

depending on their strain.

These bits are easily

detected, and provide a

molecular signature by

which one prion strain can

be compared to another. As

an added flourish, Dr

Collinge also compares the

pattern of glucose molecules

hanging on to proteins of

different strains. (In nature,

such molecular hangers-on

help a protein to do its job.)

These procedures can be

done in a day or so, but are

not yet sensitive enough to

detect very low levels of

prions that might be found

in, say, blood samples.

Dr Safar, however, has

found a way to improve his

method's sensitivity by

isolating the prions with a

substance called sodium

phosphotungstate before the

test is carried out. The same

trick could be applied to the

other tests too.

Even so, the most puzzling

(and fundamental) question

remains: How are

malformed prion proteins

able to subvert their healthy

brothers and sisters? Indeed,

Dr Safar's test complicates

the puzzle further by

confirming that a number of

slightly different molecular

perversions can be copied

faithfully in this one protein,

when none of the other

100,000 or so proteins in the

body shows the slightest

tendency to behave likewise.

For the person who solves

this problem, another Nobel

prize surely awaits.