Norman, Oklahoma
THERE is something peculiar about the tiny crescent-shaped embryo that
Leonard Zon is studying under his confocal microscope. In some ways, it looks
like a 28-day-old human embryo. Then again, it is only 24 hours old, and already
1 millimetre long, with a backbone, brain, eyes and heart, but—oddly
enough—no blood.
The embryo under the microscope is the spawn of Danio rerio
—the multicoloured zebrafish which usually lives peacefully in the tepid
waters of the Ganges. This embryo, by contrast, is playing a small but vital
role in what some say is one of the biggest gambles ever in the quest to
understand how animals with backbones develop from one-celled eggs into
creatures with tops and bottoms, lefts and rights, and a whole host of internal
organs, not to mention arms and ears or fins and gills.
Advertisement
Of all the 50 000 to 100 000 genes in the average vertebrate’s genome, many
thousands are likely to play a role in such transmogrifications. And until
recently, developmental biologists who wanted to understand how these genes
directed the early development of embryos would study fruit flies, nematode
worms, or mice. The problem is, flies and worms bear little resemblance to
vertebrates like us. Mice are much closer, biologically speaking, but their
embryos also develop out of sight in the black box of the mother’s womb.
Now, developmental biologists have a new option—zebrafish.
Development-wise zebrafish are similar to humans, but unlike us they develop
inside eggs which are “absolutely translucent” says Zon, who is a haematologist
and geneticist at the Children’s Hospital, Boston. This means that although the
fish embryos are the size of a poppy seed, practically every event in early
development can be witnessed.
The idea of a large-scale zebrafish project was the brainchild of geneticist
Christiane NĂĽsslein-Volhard at the Max Planck Institute for Developmental
Biology in TĂĽbingen, Germany. By December last year, and in the face of a
certain amount of cynicism, two teams, one led by NĂĽsslein-Volhard, the
other by Wolfgang Driever, then at Massachusetts General Hospital in Boston, had
created and catalogued almost two thousand genetically altered zebrafish. Some
of those mutant fish have spots instead of stripes, some swim in circles, some
have defective jaws. Then there are those that never make it to
adulthood—like Zon’s that fail to produce all but the most immature type
of blood cells.
But it’s still too soon to proclaim the zebrafish project a success, because
the biggest challenge is just beginning. Researchers must now find the genes
responsible for the physical changes in the mutant fish, and work out what they
actually do.
Fly gumbo
So why not stick with developmental geneticists’ old favourite, the fruit
fly, as a route to understanding embryonic development? After all, it won
NĂĽsslein-Volhard her share of the 1995 Nobel prize. She helped demonstrate
how genes lay down the body plan by marking cells for specific body
locations—the head, the mid-sections, the tail, and so on—in
everything from flies to flamingos.
But NĂĽsslein-Volhard and Driever argue that flies cannot tell us
everything we need to know about the animal closest to our hearts—humans.
The fruit fly may share its basic body plan with us, but it’s still an
invertebrate. The skeleton is on the outside, concealing an internal gumbo of
organs—muscles, nervous system, digestive tract—which bear little
resemblance to our own. Its heart is a thick tube with a single valve, it has no
lungs or liver. What’s more, even when the organs are similar-ish, the
vertebrate versions are invariably more complicated and take far more genes to
build.
NĂĽsslein-Volhard took the gamble and switched from fruit flies to
zebrafish. Nowadays, her lab is home to over 7000 tanks of fish. About 50 adult
zebrafish, which average 2.5 centimetres in length, are housed in each 12-litre
tank. The fish reach maturity at about three months, when the females start to
lay anything from three hundred to a thousand eggs a week. The researchers can
also harvest those eggs, by anaesthetising the fish, wedging her into a slit on
a sponge, and then squeezing the eggs out. As soon as the eggs are fertilised
with sperm, cell division begins. And it’s also surprisingly easy to make
genetically identical fish, or clones, by using killed sperm to trick the egg
into cell division. However embryo development starts, all major organs are in
place within two days. The embryos hatch at three days, and by five days, the
still-transparent larvae have started to swim.
The ability to spawn vast numbers of offspring very quickly is another one of
the zebrafish’s attractions. Mice also reach sexual maturity at about three
months, but they can only produce about eight pups every month or so. And
genetics, after all, often comes down to a numbers game.
Take “saturation screening”, which depends on disabling every gene in an
organism one at a time. This was the technique used to create both the zebrafish
mutants, and the Nobel prizewinning fruit fly mutants. The teams led by
NĂĽsslein-Volhard and Driever, who has recently moved his laboratory to the
University of Freiburg, added a chemical mutagen ethylnitrosourea, which
randomly damages the DNA, to fish sperm. The damaged sperm were then used to
fertilise eggs. Next, the researchers examined every one of the million or so
descendants, searching for any physical abnormalities that had occurred between
fertilisation and five days, when early development is complete.
The damaged genes may code for a structural protein, or for one of the
numerous signalling molecules that give the embryonic cells their marching
orders, telling them where to go and what to become—perhaps a nerve cell,
a bone cell, or a blood cell. So the fish might end up paralysed, or swimming in
a deranged manner, or with defective fins, or lacking blood. Finally, the
researchers identified the roughly two thousand unique physical changes, named
them—for example sloth, ziehharmonika, ikarus and moonshine
—and catalogued the different mutant fish in the December issue of
Development.
“The embryos are beautiful, completely transparent,” says Mary Mullens, who
was part of the TĂĽbingen effort and is now at the University of
Pennsylvania Medical School in Philadelphia. “Once the blood starts flowing
through the brain—around the eyes—you can see individual blood cells
travelling through which is pretty dramatic.” Mullens is now using the zebrafish
to study a group of genes that signal the development of tissues in the correct
order from the animal’s back to its belly. Others are hoping to get a handle on
human diseases. For instance, cardiovascular researcher Mark Fishman of
Massachusetts General Hospital in Boston wants to find out which genes dictate
heart size and trigger the branching of veins and arteries in the zebrafish. He
hopes they could help him to uncover some of the root causes of human heart
disease.
But first researchers have to pinpoint the damaged genes responsible for the
physical changes they have identified. The first step on the path to
enlightenment is to throw the dead mutant zebrafish into a blender, and grind up
its tissues. Then add some digestive enzymes to pull the cells apart, and crack
them open with a mild detergent so that the DNA comes spilling out.
Map reading
The next step is to find where the mutated gene lies in all that DNA. To do
this efficiently you need a map of the zebrafish genome: a series of chemical
markers (pieces of sequenced genome which may or may not be parts of the actual
genes) dotted along the 25 zebrafish chromosomes. Once the genes are located, it
is a relatively straightforward task to read their sequence, work out what kind
of protein they code for, and look for similarities between them and any one of
the thousands of animal genes already in the databases. If, for example, a
damaged version of a gene creates a zebrafish with a defective heart, and this
gene turns out to be very similar to a human gene whose role has eluded
geneticists, it’s a fair bet that the human equivalent plays some role in heart
function.
But creating the fish genome map in the first place is giving zebrafish
geneticists their biggest headache. Almost a century of fruit fly research has
gone into assembling that animal’s genome map, with its 11 000 genes, plus
another 7000 chemical markers. As the fruit fly map becomes ever more detailed
it becomes easier and easier to track down new fruit fly genes. With that
resource available, will developmental biologists invest time to create a whole
new genome map from scratch? Certainly, no one is prepared to wait a century for
the zebrafish genome map to get to the same stage as the fruitfly map. Within
the next three years, “we’d like to get 10 to 20 000 genes mapped”, says Zon. So
far the zebrafish genome map contains only 150 genes, and 1481 other
markers.
Still, almost every one of the 50 or so research teams involved in the hunt
for fish genes is contributing to the map, which was started in a low key way a
few years ago by developmental biologist John Postlethwait of the University of
Oregon in Eugene. (Postlethwait is a member of the team that back in 1988 taught
NĂĽsslein-Volhard how to look after zebrafish, as well as practically
everything that was then known about their genetics.)
One-eyed pinhead
Most attempts at locating zebrafish genes start by narrowing down the region
of interest. To do that, the researchers cross normal zebrafish with mutant
varieties, and watch how the genetic markers that are already on the genome map
are inherited with the physical abnormality. When the markers and the physical
abnormality tend to be inherited together, the damaged gene probably sits close
to the marker on the genome. The researchers then use a variety of different
techniques to sift through the region and find the gene. Just this spring, Alex
Schier and Will Talbot of the New York University Medical Center in New York,
surprised everyone by finding a brand new gene called one-eyed pinhead
on the zebrafish genome that could be important for development in all
vertebrates.
Postlethwait and his team at Eugene are also making sure that the first rough
drafts of the zebrafish map are as useful as possible. Every time a segment of
zebrafish DNA is added to the map, he checks it against existing sequences in
the human and mouse genetic databases, and marks the map accordingly when
similarities are found. That way researchers will be able to pinpoint the fish
mutants that are most likely to be relevant to human development and disease. So
far Postlethwait’s team has marked 135 human genes on the zebrafish map.
“Large pieces of chromosomes are inherited intact from [our] common
ancestors,” says Postlethwait. “It’s real surprising.” For example, one region
of human chromosome 2 contains 10 genes, and the fish equivalents are found
together on one zebrafish chromosome.
Meanwhile, Nancy Hopkins, a retrovirologist from the Massachusetts Institute
of Technology in Cambridge is refining a technique that could kick-start the
zebrafish genome map. The technique is called “insertional mutagenesis” and it
both creates new fish mutants and locates the damaged genes at the same time.
Hopkins injects healthy fish embryos at the 1000-cell stage with a retrovirus
that inserts a DNA copy of its own genetic material into the zebrafish genome,
occasionally crippling a gene. She identifies the fish with the damaged genes by
their physical appearance, then uses the genetic sequence of the viral DNA to
find the gene. The technique is still time-consuming—but the fish
researchers seem to like her approach. “It’s potentially extremely powerful,”
says Schier. “If it was done on a large scale it could really create a great
°ů±đ˛ő´ÇłÜ°ůł¦±đ.”
Hopkins doesn’t know how many new mutants she’s created; most of them are
still swimming around in tanks waiting to be identified. So far, she has
identified seven new mutants—and sequenced five unique genes, more than
any other zebrafish researcher to date. They include no arches, a gene
that when mutated creates a fish without jaws, and dead eye, which
stops normal cell division in the brain and eyes. “They just rot away,” she
says.
Not surprisingly, Hopkins is confident that zebrafish will be the key to
discovering how single-celled blobs change into multi-billion cell
animals—all without getting a cell out of place. “If we bring the right
methods to it we will find the genes required to make a little animal,” she
says. “Zebrafish is a made-to-order gene-finding device.”
* * *
Go with the glow
TO make it easier to identify at what stage in development a particular
zebrafish gene is turned on, biologist Shuo Lin of the Medical College of
Georgia in Augusta has genetically engineered zebrafish so that specific cell
types glow an eerie green when they are exposed to blue light. Lin’s zebrafish
contain a jellyfish gene for a green-glowing protein, attached to zebrafish DNA
that switches on genes in the red blood cells 12 hours after fertilisation. The
transgenic zebrafish are mated with a zebrafish carrying a mutation that affects
the blood system. By looking to see when the offspring’s blood stops glowing
green, it is possible to say to within a few minutes when a damaged gene first
starts to disrupt the development of the blood system. Similar techniques are
available for studying tissue development in mice, fruit flies and nematodes.
And Lin has also used his technique to make a zebrafish whose motor nerve cells
glow green. “This will be a paradigm for how lots of people will study their
gene of interest,” says Leonard Zon, a zebrafish geneticist at the Children’s
Hospital, Boston.
- Further reading:
Zebrafish genomics— from mutants to genes by J.
Postlethwait and W. Talbot, Trends in Genetics, vol 13, p 183 (1997) - Promoter analysis in living zebrafish embryos identifies a
cis-acting motif required for neuronal expression of GATA-2, by A. Meng and
others, Proceedings of the (US) National Academy of Sciences, vol 94, p 6267
(1997)