ONE of this century’s most ambitious biology projects has just been
completed—to the huge relief of geneticists worldwide. Over the past seven
years, about 400 researchers in 96 laboratories in Europe, North America, and
Japan have painstakingly spelt out every single one of the millions of DNA
“letters” that make up the complete instruction manual for building and
operating a yeast cell. While Saccharomyces cerevisiae is not the first
organism to have its complete genome sequenced, it is the first of the complex
organisms whose cells have nuclei.
And with that gruelling—some might say mind-bogglingly
boring—task of sequencing the yeast genome behind them, the geneticists
are slap bang in the middle of a gold rush. Everyone from yeast experts to
evolutionary biologists and even medical researchers is busily staking their
claims. “Nearly every day someone figures out a new, interesting way to use
genome data in whatever research they’re interested in,” says Robert Moyzis, a
geneticist at Los Alamos National Laboratory in New Mexico.
Biochemists are already hitting fresh seams as they sift through thousands of
genes, many of which are of kinds that have never been seen before in any plant,
animal or microbe. And with the whole yeast sequence stashed in a database, some
are running experiments on those genes, using nothing more than a
computer and an Internet connection. Geneticists who study humans are also
mining the yeast database because it can help them work out the function of
puzzling human genes. Even researchers who have a limited interest in genetics
per se expect to strike gold in the yeast genome. Evolutionary
biologists, for example, are sifting through the yeast genome for important
clues to how yeast evolved, as well as clues to the true nature of the earliest
forms of life on earth.
Advertisement
Last year scientists completed several bacterial genomes, and a whole host of
viruses were already in the databases. But the yeast sequence is special because
it is the first eukaryote cell—the class that includes all plants and
animals, including humans. Eukaryote cells are much more complex than simple
bacterial cells, since they have a nucleus and a variety of organelles, each
with its own job to do. And yeast, unlike bacteria, even reproduces sexually as
well as by asexual budding. Because of its complicated structure and lifestyle,
yeast must contain far more sophisticated instructions in its genome than those
needed to run simple bacteria or viruses.
But fortunately for geneticists, the yeast genome is tiny by eukaryote
standards, containing some 6000 genes and 12.1 million DNA base pairs, compared
to the roughly 70 000 genes and 3 billion base pairs needed to keep a human
going. And when it comes to favourite laboratory organisms, years of intensive
study have placed yeast on something of a pedestal. 51¶ŻÂţs understand its
biochemical workings better than any other plant, animal, or fungal cell, and
they can insert or delete genes at will. This makes it far easier to work out
the function of particular genes.
Helsinki to Heraklion
Despite this familiarity, sequencing the yeast genome has revealed a vast
terra incognita. Biologists have no clue as to the function of 40 per
cent of the genes they have identified, says André Goffeau of the
University of Louvain, Belgium. Half of these enigmatic genes have DNA sequences
similar to other, equally puzzling genes in fruit flies, mice, or other
organisms, but half have never been seen before. Researchers have dubbed these
genes “orphans”, because no one knows which gene families they belong to.
“They are extremely mysterious, extremely intriguing,” says Bernard Dujon, a
yeast geneticist at the Pasteur Institute in Paris. “These are the ones we want
to study first. But this is very difficult, because we don’t know what to look
´Ú´Ç°ů.”
Difficult, yes, but not impossible. Several labs are tackling the question
directly by disabling or “knocking out” an orphan gene in one yeast cell and
waiting to see what happens to that yeast and its descendants. The most
meticulously choreographed effort comes from a consortium sponsored by the
European Union, whose organisational acumen co-ordinated the sequencing of more
than half the yeast genome. In the next nine months, scientists from Helsinki to
Heraklion plan to delete a thousand of the unknown genes, one by one, in
individual yeast cells, and then measure how each knockout affects the growth
and reproductive ability of that cell’s descendants. The researchers will also
monitor when during the yeast’s life cycle those genes produce their proteins
and where in the cell the proteins go.
Following those initial screenings, the geneticists will lump the genes under
broad headings. For example, genes whose proteins straddle cell membranes might
be classed as “transport genes”, while others will be classed as “cell division”
or “stress response” genes. After this first rough sorting, the laboratories
that study specialised aspects of yeast biochemistry will try and pin down the
exact role of individual genes.
“At the end of the day, we can then turn to a specific lab and say, `It looks
like this gene has something to do with sugar transport. Please find out what’,”
says Stephen Oliver of the University of Manchester Institute of Science and
Technology, one of the brains behind the European programme, which now includes
150 laboratories.
Knockout experiments have also deepened one of the big mysteries of genetics:
the new experiments confirm that many genes may be superfluous to the cell’s
wellbeing. Researchers already knew that only about one in four genes is so
vital that knocking them out kills the organism. Now, using a remarkably
sensitive new technique they call “genetic footprinting” (not to be confused
with genetic fingerprinting), Victoria Smith, Patrick Brown, and David Botstein
of Stanford University have found that knocking out most nonlethal genes has no
detectable effect on the yeast cell. The team uses transposons, or “jumping
genes”—peculiar genes that can move around the genome—to destroy
single genes at random in a population of yeast. Then they culture all the yeast
cells, wounded and unscathed alike, so that healthier cells outcompete their
crippled neighbours. Yeast that lose essential genes disappear from the broth
immediately, and knockouts that impair a cell’s growth rate—even by a tiny
amount—fade away over the course of generations, providing the researchers
with an index of how important each gene is to the yeast’s wellbeing.
Big dividends
Yet even with such a sensitive method for monitoring a gene’s importance, the
researchers find that less than half the genes on chromosome 5—the only
one they have completely screened so far—have any perceptible effect on
yeast’s fitness. No one knows for certain why yeast hangs on to the rest of the
genes, says Botstein. Perhaps some of the apparently functionless genes may
serve brief but essential roles which have yet to be spotted, such as helping
the yeast switch from using oxygen to not using oxygen. For others, he says, he
favours the “NASA explanation—there are backup systems”—that is,
yeast (and presumably other eukaryotes, including humans) have evolved to keep
spare copies of essential genes in case a random mutation destroys one.
Molecular biologists Michael Snyder and Shirleen Roeder at Yale University,
meanwhile, are using the DNA sequences to engineer each gene so that fluorescent
tags recognise and attach to each of the proteins the genes encode. For the
first time, this gives the researchers a quick and comparatively easy way to
track when and where each of the proteins appear in the life cycle of the yeast
cell (and hence when each gene is active), providing important clues to the
genes’ functions. And yeast biologist Stan Fields at the University of
Washington in Seattle is testing whether each yeast protein sticks to any of the
other 6000-plus proteins in efforts to gather additional leads to the proteins’
roles.
The drive to learn the role of every yeast gene should pay big dividends not
just for yeast biology, but for human genetics as well. Long-lived, slow
breeding humans are a geneticist’s nightmare. Fortunately, many human genes have
yeast counterparts, and these are often so similar that the human version
inserted into a yeast works perfectly well. “It’s as though you could take a
part off a Mercedes and fix a Model T,” muses Botstein. Because of those
similarities, researchers can sometimes work out what a human disease gene does
by studying its healthy counterpart in yeast.
A few years ago, for instance, cancer specialists were racking their brains
trying to figure out how two human genes, MSH2 and MLH1,
helped trigger a common form of colon cancer. But when geneticists at the
University of North Carolina and Yale University discovered that almost
identical genes repair DNA in yeast, it was immediately obvious that the colon
cancer defect worked its mischief by preventing repair of DNA damage in cells of
the colon.
The colon cancer discovery was made a while back. But now that the sequence
of the whole genome is known a second set of benefits kicks in. “The whole is
not only more beautiful but [also] more factual than the sum of all its
fragments,” says Craig Venter of The Institute for Genomic Research in
Rockville, Maryland, whose team helped sequence the first bacterial genome, that
of Haemophilus influenzae, which was announced last summer. When their
knowledge is incomplete, biochemists and cell biologists have to cobble together
a working model of the cell’s internal machinery using only the parts they know
about—a bit like thinking a car must be propelled by the fan belt because
we haven’t discovered the drive shaft. In the 1970s, for example, biochemists
knew of only one molecule, cyclic AMP, that was capable of carrying messages
from the cell surface into its interior, and so they assigned it many jobs that
cell biologists now know are carried out by a pack of other molecules such as
calcium ions and inositol triphosphate.
With the entire yeast genome sequence to choose from, geneticists can also
study the complete set of regulatory elements that control the genes. These bits
of DNA are the key to understanding how the genome regulates the building and
operation of a cell because they turn genes on and off in response to messages
from other parts of the genome, other parts of the cell, and from outside the
cell. Genes that have similar regulatory sequences, and so are turned on and off
at the same time, may be linked in a single metabolic pathway. “People simply
sitting in front of their computers will be able to reconstruct some of the
metabolic pathways, and then say to the molecular biologists, `Can you confirm
this with an experiment?'” says Hamilton Smith at Johns Hopkins University
School of Medicine in Baltimore, who helped mastermind the sequencing of H.
influenzae.
More and more biologists are turning from their benches to their
computers—what Antoine Danchin of the Pasteur Institute in Paris calls the
in silico approach. He has already used this technique with some
success to study the genes that have been sequenced in the bacterium
Escherichia coli.
Origins of life
Danchin predicted that a gene’s pattern of “word” use—the codons that
determine which amino acids make up a protein—affects its ability to
attract the proteins that activate it. To test his hypothesis, Danchin scoured
the E. coli database and compared the codon frequencies of genes that
are active all the time with those that are only turned on for specific tasks.
Sure enough, the two groups were as different as Shakespeare is from Jonson,
confirming Danchin’s hypothesis.
Computer analyses of whole genomes should also let geneticists tackle broader
questions of genome organisation for the first time, says Moyzis. Are genes
scattered at random on chromosomes, or are the genes involved in a specific
task—say, cell division—grouped together in some way? Are
chromosomes divided into regions or “chapters” that each show a different
pattern of expression? “There’s a hope underlying this field that we’re going to
learn something about the rules of the game,” says Moyzis. “It’s amazing how
little is known about that.”
In silico genetics has also allowed yeast geneticists to start working out
how genes have evolved from one another over the course of millions of years.
Accidental copying errors have often created duplicate copies of genes, which
then become modified for a new function. These duplications, together with other
genome-wide alterations such as omissions of parts of genes or whole genes, gene
reshufflings, and the inclusion of genes from foreign species, are evolutionary
scars that betray some of the genome’s distant past, says Peter Little, a
molecular geneticist at Imperial College, London, on sabbatical at the Centre
for Molecular and Cellular Biology at the University of Queensland in Brisbane,
Australia. As genome historians hone their understanding using relatively simple
organisms such as yeast, they will become better equipped to work out the
evolutionary history of more complex species such as humans. They may even be
able to get a glimpse of the earliest life on Earth.
Geneticists have now sequenced complete genomes for various bacteria as well
as yeast—that is, representatives of two of the three great branches in
the tree of life. The complete genome sequence for Methanococcus
jannaschii, the first representative of the third branch, the Archaea, is
about to be published, by evolutionist Carl Woese of the University of Illinois
at Champaign-Urbana. By comparing the genetic tool kits of these disparate
organisms, Woese hopes to deduce which genes were present in their common
ancestor, back when the tree was just a seedling. This will give clues to the
nature of that organism, and the earliest forms of life on Earth. “If biology is
to be a full science, it must account for the origin of life on this planet,”
says Woese.
Evolutionary biologists also hope the complete genome sequences will help
settle debates that have often split their field over the past
decade—including, for example, whether eukaryotes, such as yeast and
humans, are indeed a third major branch of the tree of life or a hybrid of the
two other branches created by an ancient symbiosis between a bacterial cell and
an archaean cell, as biologists Brian Golding and Radhey Gupta of McMaster
University in Hamilton, Ontario, propose.
But although decoding the complete yeast genome promises to answer questions
about evolution, yeast biochemistry, gene regulation, and some human diseases,
it can’t help sort out the major conundrums raised by complex organisms like
worms and humans—such as how genes help organise cells into a
multicellular whole, or the nature of the genetic instructions that run a brain
and a sophisticated nervous system. To tackle those issues, researchers must
wait for the complete genome sequences of a tiny roundworm called
Caenorhabditis elegans— expected in 1997—and of the human
—expected around 2005.
And in that respect the gold rush triggered by the completion of the yeast
genome is just a harbinger of even greater genome research frenzies to come.
Which is exactly as it should be, says Venter—after all, “these genomes
are the legacies we’re leaving the next thousand years or more of science.”
