By David N. Leff

Pet stores usually sell their inch-long, brightly striped zebrafish (Danio rerio) by the breeding pair. Nancy Hopkins has bred and pampered 36,000 of the little fish, but they're not for sale. Rather, she keeps them as founding parents, to beget children, grandchildren and great-grandchildren finny specimens.

Her myriad third- and fourth-generation zebrafish are not for petting but for gene analysis. Hopkins, a molecular biologist at the Massachusetts Institute of Technology in Cambridge, says that if she and her co-workers can identify which D. rerio genes are crucial for normal fish development, "we will have added enormous value to the Human Genome Project."

This international undertaking aims to sequence the entire genome of Homo sapiens in the next few years. However, Hopkins points out, "While identifying the sequence of all 100,000 human genes, it cannot tell researchers what those genes do, or which of them work together to develop a normal heart or a healthy nervous system." She intends her generations of cloned, mutant zebrafish genes to fill that gap.

D. rerio's genome is about 1.6 x 10⁹ base pairs (1.6 billion) long, which makes it roughly half the size of the human genome. The gene count in humans is estimated at about 100,000. Zebrafish are probably similar.

Molecular geneticist Adam Amsterdam, a post-doctoral associate in Hopkins' lab at MIT, pointed out that "Fertilization in D. rerio is external. This, plus the embryo's transparency, is why people study zebrafish; they can watch everything of their development from the one-cell stage on to adult. By the time 24 hours have elapsed since fertilization, you have a pretty recognizable organism. It already has eyes, ears, a brain, and axial structures such as a notachord and neural tube, as well as somites that eventually will become the muscles and skeleton of the fish. The heart begins to beat around 32 to 36 hours after fertilization. Embryonic development," he went on, "ends at five days, by which point the fish is swimming around, looking for food, ready to go. But It's not sexually mature until about three months of age.

"Broadly speaking," Amsterdam said, "our goal is to identify and clone as many genes as possible that are required for the embryonic development of the zebrafish, in the course of these five days."

Wholesale Mutagenesis Yields Retail Mutants

Hopkins is senior author, and Amsterdam lead author, of a paper in the October 15 issue of Genes & Development - a journal of the Cold Spring Harbor Laboratory Press in New York. Their report bears the title: "A large-scale insertional mutagenesis screen in zebrafish."

To insert mutagens wholesale into their zebrafish founders, Amsterdam told BioWorld Today, "We start with the mouse Moloney retroviral vector (RVV), because it can't replicate. But there's a little trick that we do to it, called pseudotyping. For example, the mouse RV that we're using would not normally be able to infect fish cells. There's a specific protein on the surface of the virus that interacts with a specific protein on mouse cells, so it can't infect zebrafish cells.

"What we do is co-express sequences of that Moloney RNA virus with the envelope protein of vesicular stomatitis virus, which is capable of infecting all kinds of cells from all different species, including vertebrates. The first thing that RNA virus has to do is reverse transcribe, to make DNA out of itself. And that DNA copy of the RNA virus, called a provirus, is what gets incorporated into the genome.

"The reason we use a RVV at all," Amsterdam continued, "is that it makes a copy of itself, and inserts it into the chromosome of cells it's infected. That's the goal we're trying to achieve - to disrupt zebrafish genes by sticking something foreign into their genome.

"We inject this virus into very early-stage embryos," Amsterdam recounted, "when they don't even look like a fish yet. They're just a hollow ball of blastula cells, about a millimeter in diameter. There's just enough room to get the liquid vector into the middle of the ball; we shoot the virus in through a needle.

"Then the RVV infect many different cells in these founder-parent embryos, but those fish grow up okay. What we're counting on is that some of those cells we infect will contribute to their progeny's germ line - either the sperm or the egg. So when those second-generation fish grow up, we mate them and wind up with fish that now have whatever insertions were in the germ cells of the founders - the ones injected - which are now inherited in a small proportion of the next-generation fish.

"Then what we need to do to find the mutations is breed those mutated progeny to homozygosity. We are looking essentially for recessive mutations - inherited from both parents - which we can detect only in the homozygous state. In total we breed for three generations, and given that it takes three months for a fish to be sexually mature, that takes roughly a year, from founder injection to third-generation mutation."

Amgen Seeks Piece Of Zebrafish Action

Those mutant fish display anomalies affecting many different parts of the embryo. These include generalized growth misbehavior, and specific errors of pigmentation; the eyes; the ears, the liver, and so on. En route to their goal of 1,000 mutant genes within the next two years, Hopkins and her co-authors have isolated 45 so far. One, for example, results in zebrafish that fail to grow to normal size. They remain tiny normal replicas of perfect adults. Another anomaly produces fish whose fins never stop growing.

These preliminary outcomes have not escaped the notice of Amgen Inc., of Thousand Oaks, Calif. The company is partially funding Hopkins' zebrafish program, and doing some of the team's DNA sequencing.

"Looking into the future, as Amgen is," Amsterdam opined, " I think their interest is in finding soluble factors that could for example be used after chemotherapy, which can kill certain cell populations. Such factors could renew and stimulate the production of those types of cells. I think their thought is that if we can find genes that are important in development for establishing those cell populations in the first place, that those might be potential products, that those genes would be the ones that you would want to try to use in patients."