© Mark Smith/Photo Researchers

The slim, striped zebrafish, one of the newest and wettest members of the model organism family, has attributes that researchers look for: size, fecundity, and low maintenance costs. Toss in a couple more assets--its visible internal development, and its genetic similarity to mice and humans--and this fish's return-on-investment increases. But it arguably would not have gained its formidable reputation had it not responded so well to new techniques.

Studied initially to understand development of the cardiovascular and neurological systems, these popular aquarium pets have been subjected to large and small screens, fluorescently labeled and fed quenched fluorescent phospholipids; their embryos' nuclei have posed for high-resolution time-lapse pictures.¹ Researchers have developed knockdown fish technologies and also are using techniques that previously belonged to the dominion of plant scientists. Forward genetics techniques propelled the diminutive teleost into the hearts and labs of life science researchers; reverse genetics was recently added to the aquatic tool box.

Experimental results are remarkable. Mark Fishman, head of Novartis Institutes for Biomedical Research, Cambridge, Mass., says that genetic screens have revealed the logic behind the design of the heart ventricle and its endothelial-lined circulation. Behavioral screens have identified gene mutations involved with retinal degeneration. Biochemical analysis of blood, which was taken from tail snips, led to the isolation of fish defective in clotting.

Many zebrafish mutants were first identified by observing the color and number of blood cells flowing through their veins, or by noting gross defects such as brain and heart deformities. And since zebrafish share many structures with their higher (but opaque) vertebrate cousins, they can model both development and pathology. "The body plans of the vertebrates are pretty similar," says Perry Hackett, chief science officer for Discovery Genomics, a zebrafish contract research organization.

Mutant screening is now aided by fluorescent labeling, creating transgenic lines with neural-specific or blood-specific GFP expression, for example, or by injecting dyes into the heart or other developing organs. Mutations that exist in the labeled systems then can be visualized as a loss (or gain) of fluorescence in the mutant fish, as compared to the wild type.

These and other screens show that developmental defects can be tracked in the transparent embryo by light microscopy, allowing visual access to organ primordia and the developing nervous system as the fish matures in a Petri dish. "Thousands of mutants have been generated in every organ system you can think of," says Harvard University's Leonard Zon, whose lab uses the zebrafish model to investigate cardiovascular disease and cancer. Among these, says the University of Oregon's Charles Kimmel, "you can see individual neurons in the brain." Moreover, says Zon, more tools are coming. "The history of the zebrafish has not been to rest on its laurels, but to boldly go and create newer and newer technology."

HUMBLE BEGINNINGS The late geneticist George Streisinger, Kimmel, and colleagues spent years in virtual isolation at Oregon, establishing breeding and mutagenic protocols that would serve as the foundations of zebrafish research.² Cecelia Moens, once a murine researcher, joined Kimmel's lab to search for genes that caused brain defects--something that she couldn't do in mice. "It wasn't until the people in Oregon had shown the development of the zebrafish embryo was both accessible and similar to the development of other vertebrate systems; and that you could actually mutagenize and find mutations in zebrafish," says Moens, that other scientists began to notice. Her lab at the Fred Hutchinson Cancer Research Center is now working with reverse genetics to discover the relationship among genes responsible for architecture of the hindbrains.

Moens continues: "Those two things together made people like Mark Fishman and Janni [nobelist Christiane] Nüsslein-Volhard believers; to think that you could use this system in a large-scale way to try to identify all the genes that are involved in controlling early development." That belief led Fishman and his former colleagues at Massachusetts General Hospital, and Nüsslein-Volhard's team in Tübingen, Germany, to undertake the massive screens, begun in 1993 and published in 1996, that have cemented the zebrafish's place as the newest of major model organisms.

Of course, money funds research, which encourages more research. The trans-NIH zebrafish genome initiative began in 1997 to clone and generate physical and genetic maps of the genome. In 2000, Artemis Pharmaceuticals completed a second major Tübingen screen. And, in 2001, the nonprofit Wellcome Trust Sanger Centre published a preliminary draft of the zebrafish genome.

Tammy Irvine, Rear View Illustrations

ON THE BIG SCREEN Fishman wants to understand how the cardiovascular system develops--beginning with a precursor cell's initial commitment to a lineage--and how it may have evolved. "Certain higher-order structures of the cardiovascular system are 'new' to the vertebrate, including both a high pressure-generating chamber, the ventricle, and an endothelial-lined circulation," he says. "The organizing logic of both these new modules, the ventricle and the endothelial-lined circulation, have been revealed through the genetic screens."

Ron Plasterk's group at the Hubrecht Laboratory, Netherlands Institute for Developmental Biology, Utrecht University, has constructed parallel sperm and DNA libraries from hybrids between mutagenized males and wild-type females. After nested PCR amplification screening for a mutation in a gene of interest, they recovered and bred "target-selected" zebrafish. "The method is, strictly speaking, not targeted-gene inactivation ... since the mutagenesis is random," Plasterk explains. But the "bottom line is that gene targeting is now possible in the zebrafish." They used the technique to generate and characterize rag1 recombinase knockout lines. The paper detailing the results claimed that the 2,679 samples created are, in principle, "comprehensive enough that most genes should be represented by at least one null allele."³

Plasterk's and other zebrafish laboratories are using several related techniques to make knockout animals by generating and detecting mutations in selected genes. TILLING (targeting induced local lesions in genomes), originally described in 2000 for Arabidopsis screening, is one of the newest. TILLING uses denaturing high-performance liquid chromatography coupled to sophisticated software to detect slight differences in melting temperatures between PCR-amplified wild-type homoduplex and wild-type mutant (mismatched) heteroduplex DNA. Researchers also are exploring combining TILLING with the celery-derived endonuclease (cel I) capable of recognizing and cleaving hetero-duplex DNA.

Steven Farber's group at Thomas Jefferson University in Philadelphia fed zebrafish a quenched fluorescent phospholipids, ped6, to probe the workings of the digestive system. This phospholipase A2 substrate is cleaved in the intestine and transported to the gall bladder by way of the liver, with fluorescing metabolites illuminating pathways along the way, demonstrating "that ped6 metabolism corresponds to established mechanisms of lipid processing in mammals."⁴

The researchers then used ped6 for a pilot screen on the F2 larvae of mutagenized fish. They identified two mutants defective in lipid metabolism. One of them, fat free, has a morphologically normal digestive system that more conventional screens would not have detected. Nonetheless, the mutant had severely reduced phospholipid and cholesterol processing.

IN THE RING Scientists also have figured out how to make knockdown zebrafish. Antisense morpholinos (nucleotide analogs that natural enzymes do not degrade) are injected into the yolk of 1-4 cell embryos and spread throughout the developing fish, inhibiting production of the protein of interest.

"This has made the field available to a lot of investigators; they want to find out what phenotype would happen if their particular gene were knocked down," notes Zon. A quick Medline search bears him out: In 2002, 41 papers were retrieved using the search keywords zebrafish and morpholino. The papers describe a range of phenotypes, including kidney disease and skin abnormalities. The technology, introduced at the biennial zebrafish meeting three years ago, has become widespread enough that last year's meeting featured a talk about using modified peptide nucleic acids as an alternative to morpholinos for creating knockdown zebrafish.⁵

"These are really exciting times in the fish field," Zon says. From TILLING and morpholinos to the imminent introduction of homologous recombination, zebrafish researchers are no longer confined to forward genetics. "A lot of new technology has come about over the past four years.^6,7 For the first time it's going to be possible to do reverse genetics." Discovery Genomics' Hackett said that Streisinger went looking for a "phage with a backbone" in the late 1960s. One wonders if he knew exactly what he reeled in.

Josh P. Roberts (tcwriter@msn.com) is a freelance writer in Minneapolis, Minn.