¿ìè¶ÌÊÓƵ

If we must pay billions of dollars for a map of the DNA in our cells,
what’s in it for us? Will biologists net a few curiosities for themselves,
like moon rocks to exhibit in a museum? Or, as its supporters like to say,
will the human gneome project change our very way of life? In fact, the
practice of medicine, indeed, how we think about medicine is changing because
of the techniques that the genome project is perfecting. A good place to
view the metamorphosis is in the laboratory of Francis Collins, a physician
and geneticist at the school of medicine of the University of Michigan in
the US. Collins treats children with cystic fibrosis (CF), a hereditary
disease that leads to chornic and eventually lethal lung infections. He
also treats people with neurofibromatosis (NF), a disease that causes tumours
sometimes so disfiguring that it was once thought that the Elephant Man
of 19th-century fame had it.

But Collins is also adept at the abstract calculus of molecular genetics.
In 1989, his team at Michingan’s Howard Hughes Medical Institute tracked
down the genetic defect that causes cystic fibrosis (in collaboration with
Lap-Chee Tsui and John Riordan at the Hospital for Sick Children in Toronto,
Canada). A year later, first collaborating with Ray White at the University
of Utah in Salt Lake City and then in competition with White, Collins’s,
team unearthed the gene responsible for NF. Only six genes for inherited
disease have been exactly located, although many markers have associated
with disease genes have been pinpointed. Collins’s two qualify him as an
ace among gene hunters. ‘When I go into this area a few years ago, the idea
of finding a gene by positional cloning was frightening,’ Collins recalls.
Positional cloning means finding a gene without knowing the protein it makes,
or anything about its DNA structure, by using its map position to indentify
it. It is a time-consuming exercise in juggling pieces of the genome using
the same techniques gene mappers employ. ‘We were lucky,’ Collins says of
the search for the CF and NF genes. Both diseases are fairly common, CF
occurring in one of 2500 people in the West and NF in one in 3000. That
provided geneticists with many families who had a history of the diseases.
By comparing the DNA of patients with their unaffected relatives, Tsui in
1985 traced the CF gene to a section of chromosome 7. Then White in Utah
and Robert Williamson of St Mary’s Hospital Medical School in London found
‘markers’: small pieces of readily identifiable DNA, that are almost always
inherited with the disease, and so must be very close to it. That left Collins
and Tsui in a daunting but unmanageable region of 1.6 million bases, the
chemical units of DNA, in which to search for the gene.

In 1987, the gene for NF was traced to a larger but unusual section
of chromosome 17. After that, the true toil of molecular biology began:
cutting and pasting lengths of DNA to walk and jump along the chromosome,
stopping now and then to examine fragments that might contain the crucial
gene. It took two years.

If finding the CF and NF genes was painstaking, tracing more complex
diseases would be Herculean with current techniques. Geneticists suspect
that many diseases are cuased by defects in several genes, which might be
scattered across the genome. For example, in a family with a history of
breast cancer, several mutations might be at work. Other differences in
a family’s genes may be harmless variations, known as polymorphisms. It
could take 10 times the effort Collins and White made to find NF to locate
such disease genes. ‘Your genetic mapping ability starts to blur out,’ says
Collis, when more than one gene is involved.

The burst of gene discoveries over the past five years – for CF, NF,
retinoblastoma, Wilms’ tumour, a form of colon cancer and muscular dystrophy
– will dry up unless better methods to map come along. ‘Breakthroughs in
technology are going to be part of the genome project effort, Collins predicts.
‘The fact that there is a genome project will stimulate that technology
to appear.’

The plan for mapping and sequencing the human genome has performed its
own mutation since the mid-1980s. Then, the Department of Energy proposed
a massive effort to sequence it all. But the National Instittues of Health
in Bethesda, Maryland, has wrested control of the project and laid it at
the feet of James Watson, who shared a Noble prize for the disvoery of the
structure of DNA and is now first tsar of the genome project. The DEO remains
the ‘lesser equal’, concentrating on designing the new technology that Collins
eagerly awaits.

In the meantime, the NIH will underwrite many of the medical applciations.
How much of the genome project’s budget should be spent looking for specific
disease genes provoked some of the bitterest debate over the project. Medical
geneticists argued that the areas around disease genes should take priority.

Neither side has won, but the current compromise has raised the status
of medical applications in no small measure because Congress must find $200
million a year for the project and politicians need cures to offer their
taxpaying constituents. Many medical geneticists, however, are de facto
mappers, tracing the genome as they search for disease genes. Many now see
merit in looking beyond disease genes.

Says Collins: ‘I think it makes a lot more sense to say at the outset
that most of the genome will have interesting genes for something. We don’t
know where they are going to be ahead of time.’ He argues that a bevy of
separate miniprojects, each devoted to a particular disease, will be less
effective and more expensive than a coordinated project that encourages
the application of the newest technology.

Charities fund basic research

Collins’s laboratory at Michigan focuses almost entirely on diseases
– CF, NF and Huntington’s disease – and on genes coding for haemoglobin,
and his is not among the four labortories funded by the genome project’s
section on medical genetics. He gets backing instead from medical charities.

As symbiosis of sorts has emerged between the pure mappers and the gene
hunters. Each group has created tools along the way that help the other.
Collins, for example, holds a patent on a technique known as ‘chromosome
jumping’. It was the trick which Collins and his colleagues used to traverse
unknown genetic territory and that originally drew White into his short
collaboration with Collins in the NF hunt. Some 50 laboratories now use
the technique.

Mappers have in turn contributed to the armamentarium of medical scientists.
Yeast artificial chromosomes (YACs), for example, were developed by mappers
at Maynard Olson’s laboratory at Washington University in St Louis. They
consist of a yeast chromosome with a large piece of human or other foreign
DNA attached. The technology allows large segments of DNA to be manipulated,
making quick work of an ordered NF team – Lone Andersen, Margaret Wallace
and Doug Marchuk – to map the region around the genes.

With Charles Cantor, the scientific architect of the DOE’s mapping project,
Olson also has proposed creating signposts for mappers called sequence tagged
sites (STS). They suggest that every scientist who isolates a significant
clone in the course of research should denote a pair of short, flanking
segements of the clone and publish their position and sequence. From this
information, scientists anywhere can find the sequence in their own store
of DNA and reproduce it without having to wait by the postbox for biologcal
samples to arrive.

Steven Warren, a medical geneticist at Emory University in Atlanta,
Georgia, employs STS in his search for disease genes on the X-chromosome,
one of the two sex chromosomes. He supports the genome project because his
areas on the X-chromosome is ‘gene dense’ and is suspected of being the
source of many inherited conditions, including genes for a variant of muscular
dystrophy, manic depression, and certain skeletal abnormalities. ‘It would
be dificult to map the region without the genome project because study groups
(who decide which research to fund) would probably say it was unfocused,’
says Warren. ‘They want you to go after a disease. But when a map (of the
region) is completed, you can position a genetic disease between the markers.’
Warren is one of four researchers already funded by the genome project’s
section on medical genetics. The others include David Housman of MIT in
Cambridge, Massachusetts, to study Wilms’ tumour; Ray White at Utah, who
is examining adult polycyctic kidney disease and NF; and James Gusella at
Massachusetts General Hospital in Boston, who is looking for the gene for
Huntingdon’s disease.

Slow progress towards a cure

Gene therapy will begin later this year on children whose gene for an
enzyme, adenosine deaminase, does not work, thus fatally weakening their
immune system. But gene therapy may also soon be used not merely to replace
defective genes, but as a new sort of drug therapy. The idea is to insert
particular genes into genetically normal cells so that they produce a therapeutic
protein. For instance, genes could be inserted into the cell walls of blood
vessels where they can forever pump out proteins to fight tumours to scour
plaque from arteries. Viagene, a company in San Diego, California, may soon
test a therapy for HIV infection in which a gene taken from the virus is
inserted into the DNA of infected people. In cells of the immume system,
it will produce viral antigens to provoke the body to mount a cellular immume
response that, theorectically, could wipe out the virus.

Mapping will help to lead the way to these genes. Some debate continues
on what to record along the way, however. Medical geneticists would like
mappers to add complementary DNA to their agenda. When DNA is transcribed,
or red, to make a protein, messenger RNA is created as an intermediary step.
Researchers can extract messenger RNA from a cell and create a complementary
copy of the original DNA, and then use this as a probe to find the gene’s
location. If they get a match, they know there is a gene; if not, it is
junk DNA, which makes up about 85 per cent of the genome.

Finding a disease gene does not automatically lead to a cure. It takes
time to determine what a gene does, or does wrong, in the case of most hereditary
disease. ‘There is no cure for NF,’ says Collins. ‘In fact, with every genetic
disorder that goes this route, we are going to have an uncomfortable window
that may extend over decades, when we can diagnose but not treat.’ For example,
Collins and White found that different mutations in several NF patients
caused their disease. Collins surmises that each family will have its own
particular collection of disease-causing mutations. With CF, the original
mutation causes only about 75 per cent of the cases; about 50 other mutations
have been found that also cause CF, and no single one is responsible for
more than a few per cent of the remaining cases.

The variety of genetic missteps leading to the same endpoint makes any
standardised form of gene therapy extremely complicated. Nonetheless, finding
the gene for a disease points medical scientists toward the protein it makes,
possible leading to a drug or other type of therapy. Last month, for example,
Collins and White discovered that the NF gene is very similar to another
gene that creates a protein, GTPase-activating protein, or GAP, which controls
cell growth. Making it easier to cross-check one gene against libraries
of other genes for similarities is part of what the genome project hopes
to accomplish. So the search for genes, either to fix them or harness them,
will accelerate, driven by both bedside physicians looking for cures and
mappers sent out, like Magellan and Columbus, by the human genome project.

From motorcycles to molecules: gene sleuth in the fast lane

An editor of an Italian motorcycle magazine recently telephoned the
press office of the University of Michigan, wanting to get in touch with
Francis Collins. The journalist was not impressed so much by Collins’s innovative
techniques such as chromosome jumping as by the photograph of Collins in
an issue of the newspaper USA Today. The tall, sandy-haired gene hunter,
leather-jacketed with helmet tucked under his arm, was standing next to
his Yamaha motorcycle. Could they have an interview? he asked.

Collins is probably America’s most quoted medical geneticist. There
are others of equal skill who run in the same elite pack of the gene hunters,
but Collins is the one the talk show hosts want, especially those that showcase
the likes of transvestites, fire-walkers or former drug dealers. They like
him for the same reason journalists do: he describes molecular biology in
a way they can understand.

Collins attributes his knack for metaphor and analogy to his patients.
‘I counsel people with genetic diseases every Friday. I explain sophisiticated
facts to people with a grammar school education, and I have failed them
if they go out shaking their heads.’ He also credits his upbringing: his
mother was a playwright and his father taught literature at a small college
in the Shenandoah Mountains near Staunton, Virginia. Collins chose physical
chemistry and motorcycles, however before turning to medicine. He settled
finally on genetics because it is based on principle rather than detail.
‘That means you can take basic principles and deduce a lot about life without
having to memorise a lot of facts, which is what I didn’t like about biology
at first.’

Moral principles also preoccupy Collins. He is often reminded how much
potential the information he is uncovering has for harm as well as healing.
‘People are extremely anxious to have fantastically productive children.’
Even now, he says, there are already people who come to him in genetics
clinics asking for an amniocentesis to find out the sex of the fetus. ‘If
it’s not what they want they would terminate the pregnancy and try again.
I find that abhorrent. It violates every reason I went into this business,’
Collins says. ‘Sex is not a disease, sex is a trait.’

Consumers might want to do the same with inherited qualities to make
their children smarter or more athletic. ‘I hate to say it but it sounds
like we need some oversight over the application of this technology to that
sort of trait selection.’

By the same token, genetic research should not be halted in the meantime,
Collins argues. ‘But we have to have a public that is better educated about
genetics,’ says Collins. ‘It’s a real disaster now.’ The US already needs
many more genetic counsellors to guide parents, expecting couples and anyone
else taking a genetic screening test through the intricacies of inheritance.
For this reason, Collins hopes that the screening for the cystic fibrosis
gene that he discovered is not extended, at least for now, beyond people
from families with a history of the disease.

Collins’s team is in great demand by others in medical genetics, and
he ecnourages intense collaborations; genetics is moving too quickly, he
says, not to share information that might soon help patients. When he collaborated
with Lap-Chee Tsui of Toronto on the search for the cystic fibrosis gene,
the two regularly met at a halfway point – London, Ontario – to compare
progress. Collins’s laboratory in the University of Michigan’s Division
of Medical Genetics (which Collins heads) also participates in the collaborative
research agreement of the Hereditary Disease Foundation, an unusual experiment
in which six research laboratories agree to swap information freely and
immediately.

Groups respresenting victims of hereditary disease speak almost reverentially
of Collins. His secretary calls him Francis and his assistants fearlessly
confess their failed experiments, to be led to their own solutions with
some gentle Socratic probing. If Collins ever does anything that really
annoys people, it is probably when he admits to being a serious Christian.
‘That gets me in hot water sometimes,’ he acknowledges mostly with other
scientists. He decided 10 years ago, at age 30, to examine religion closely
for the first time in his life. When he became a serious believer, some
colleagues assumed he had committed ‘intellectual suicide’.

‘It seems to be a very well kept secret,’ replies Collins, ‘but there
is no major conflict between those two sets of belief systems .. and by
the way, in case you are wondering, I do believe in evolution.’

How they found the gene for neurofibromatosis

Francis Collins compares the search for a disease gene to finding a
burnt-oug light bulb in a house with no address in an unkown street in an
anonymous city somewhere in the US.

Actually, medical geneticists in search of the NF gene had some idea
of where they are going. For example, Collins and Ray White at the University
of Utah knew from studies of families that the NF gene lay on chromosome
17. The next step was to position it in relation to markers that were inherited
with the disease, because they are physically near it along the length of
DNA.

White and Collins knew by 1988 that some poeple with NF have two translocations
on chromosome 17. Translocations are mutations where pieces of DNA switch
positions, sometimes, between chromosomes. If a translocation occurs within
a gene, it can disrupt its behaviour, leading to disease. The translocations
in people with NF ‘pushed the limits of coincidence’ that they were benign
or unrelated somehow to the disease gene, recalls Margaret Wallace of Collins’s
team.

So the area to be searched shrank from a stretch of DNA that, at about
five to six million base pairs, seemed to stretch into invisibility, to
a region between and around the translocations, of only about 100,000 base
pairs. The translocations themselves lay only 60 base pairs apart.

The researchers had already started ‘walking’ down the DNA towards the
NF gene by cutting up the DNA between and flanking the translocations with
various restriction enzymes. Each enzyme cuts DNA when it recognises a certain
small sequence of base pairs. One way to ‘walk’ is to use enzyme A to cut
a length of DNA into 50 fragments, or clones, of varying length, then use
enzyme B on the same stretch that might divide it into 40 pieces. If the
tip end of a clone from the A library sticks, or hybridises – remember that
a single strand of DNA seeks out and binds to its original complementary
strand – to a clone from the B library, they must have been contiguous.
Think of two copies of the same book, one divided into 10 chapters, another
into 15. Overlaps in the texts of the two versions will occur between contiguous
chapters.

Collins and his colleagues also used yeast artificial chromosomes. These
allowed them to manipulate large segments of DNA, like concrete sections
rather than brick by brick, to construct their map. Both walking and YACs
have been honed by mappers as well as medical geneticists. Doug Marchuk
of Collins’s team used YACs in a novel way, by putting the region suspected
to contain the NF gene into a YAC andusing it to probe cDNA, hoping for
a match.

Chromosome jumping moves much faster than walking, however. Collins
helped to invent the technique while at Yale University, and holds a patent
on the process. One can cover much more territory by ‘jumping’ over large
sections, as well as leaping over bits that cannot be cloned. Essentially,
one creates a library of clones with one enzyme cutter, and then joins the
ends of the clone together to form a circular piece of DNA. If the circular
piece is cut with another enzyme, a clone that includes what used to be
the ends will be selected out and used to map the region. In so doing, Collins’s
team jumped across a sizeable chunk of chromosome 17.

The gene must lie somewhere near the translocations, they reasoned.
They got a fortuitous clue from Art Buchberg at the National Cancer Insititute,
who had found a mouse gene and sent it to Collins asking if it was a leukaemia
gene. It wasn’t, but to Collins it was something even better: a gene that
matched a section, perhaps a human gene, between the two translocations.

The mouse theory foundered, however. The gene found in NF patients did
not differ from that of healthy people, nor did it fall across either of
the translocations. But Collins’s team found other candiate genes in the
region by comparing pieces of DNA to libraries of cDNA expressed in peripheral
nerve tissues and white blood cells. They got two matches, which turned
out to be pieces of the one huge gene. It was so big that ti had to cover
both translocations in the NF patients, says Wallace.

Another clue came from ‘zoo blot’, an assay in which human DNA is compared
with libraries of genes from various organisms, from bacteria to primates.
Genes tend to be conserved among species; when a piece of mysterious human
DNA matches up with an animal gene, it is a fair bet that it’s a human gene
and not ‘junk’. They got a match with mice and rats.

The genome project’s plan include mapping and sequencing several organisms,
including the mouse, which will given gene hunters more territory in which
to roam. The final clue, however, came from a patient. Collins knew that
the gene was large and mutated often, and thus would contain many mutations,
many of which would be harmless. How to pick out the guilty one? By comparing
the DNA of a patient with the disease against his or her unaffected parents,
he reckoned he might spot a new mutation that caused disease. He found one,
an insert of 500 base pairs, nestled within the suspected gene.

White had used a slightly different strategy to track down the gene,
although the mutations that caused disease in his patients were different
from Collins’s patients. (The fact that many different mutations can cause
NF will make drug design or gene therapy difficult). Both published their
results the same week.