One of the best articles that I have read on genomics and proteomics. This Financial Times September 2, article is in laymans terms.
BODY AND MIND: The gene is done. Now fold your protein: WEIRD SCIENCE:
Jerome Burne plots a path through the post-genomic world of mutant mice and
designer drugs
Financial Times, Sep 2, 2000
Completing the humangenome project may have been a giant step for man but
it's very bad news for mice. The next big thing is a European project to induce
mutations chemically in 40,000 of them. Thismeans injecting them with the
gene-damaging drug ENU so their offspring suffer defects such as cataracts,
abnormal limbs, deafness and obesity.
The aim of the experiment, run by research groups in Germany and the MRC
Mammalian Genetic Unit in Oxford, is to help answer the big question thrown up
by the genome project: "What do all these newly found and mapped genes
actually do?"
Mice are used extensively in genetic research because they share a large
percentage of their genes with us. Knowing that a faulty gene in mice causes
problems with breathing, for instance, means that the same gene in humans is
almost certainly active in our lungs as well.
Mass mutation of mice, however, is just one of several large-scale, post-genome
projects which promise to change the face of medicine in the coming decades.
More immediate results can be expected from another hot genetic topic - designer
drugs. The new anti-HIV drug, Ziagen, due to be launched soon in the UK, will
come with its own DNA testing kit, to check that patients have the right genes to
benefit from it.
Nor is it just Aids patients who will be matching their genes to their medication.
Do you know which one of the three versions of the gene CYP2D6 you have?
Almost certainly not, but you soon will. This is the gene that codes for an enzyme
that breaks down about 50 widely prescribed drugs, among them tricyclic
antidepressants, codeine and beta blockers.
One version, carried by about 1 per cent of white Europeans, acts unusually
quickly, another carried by 6 per cent, is very sluggish. Having an unusual version
of the gene can produce dramatic results. While a poor metaboliser might only
need 20mg of the antidepressant nortriptyline, fast responders can handle
500mg.
Fast metabolisers can suffer the effects of a morphine overdose from codeine,
while poor metabolisers get no benefit from it at all, although they may collapse
from taking the blood- pressure-lowering drug debrisoquine.
A company called GeneSolve, a partnership between Dundee University and the
recently privatised Laboratory of the Government Chemist, is planning to market a
chip for use in every GP's surgery which will tell you which version of CYP2D6 you
have, allowing personalised prescriptions.
All this new technology has had a nasty side-effect for drug companies, however.
In order to show how much better designer drugs are, they have had to come
clean about the shortcomings of their existing product line.
For instance, large percentages of patients don't respond to all sorts of widely
prescribed drugs such as cholesterol-reducing statins (30 per cent), beta
blockers (35 per cent) and tricyclic antidepressants (50 per cent).
What's more, patients, especially those taking psychiatric drugs, may have to wait
months or even a year to discover if they work at all. A US study has found that
106,000 patients die and 2.2m are injured each year by adverse drug reactions at
a cost of Dollars 100bn. In Britain, an estimated 6 per cent of hospital admissions
are caused by prescription drugs and the official figure for serious drug reactions
is 20,000, but experts believe it could be 10 times that.
But it is not just designer drugs that are going to improve the situation; the next
really big thing is proteomics - the study of proteins. One of the biotech giants,
Structural GenomiX, is planning to spend up to Dollars 500m over the next five
years determining the shape of just 5,000 proteins. Compared with proteins -
what genes actually make - sequencing the genome was a doddle.
There are several times more proteins than genes. While DNA has an alphabet of
just four letters, proteins are built from a pool of 20 amino acids. Each of our 250
cell types produces a different pattern of proteins which change over time, and at
the moment we have no idea what most of them do. Proteomics holds out the
promise of developing far more precise disease markers and drug targets. At the
moment, only 500 proteins are affected by all our drugs. That could rise to 10,000.
One of the biggest mysteries about proteins is that each one is folded into a
complex 3D shape. Even if you know its amino acid recipe, there is no way at
present of predicting what shape the final protein is going to be. Uncovering the
shape of a protein, known as structural genomics, can give a clue to what it does
and make it easier to design drugs to affect it.
As a measure of the complexity involved, IBM has started building a computer,
called Blue Gene, which it is said will be 500 times faster than anything available
today. It will attempt to predict how a protein will fold by calculating the atomic
forces acting on the amino-acid chain every quad-rillionth of a second. It is
expected to take a year to simulate the folding of a single protein, something the
body does almost instantaneously.
Mice, no doubt, will eventually be helping out on this one as well. |
