An Interview With The Zebrafish

Forget rats, mice, and guinea pigs. Here's the ultimate research subject, and it's prepared to reveal everything about itself.
May 15, 2013

Hello, I'm Danio. I know it’s crowded, but come in. I started showing up in places like this in the 1980s, and now I come here by the tens of thousands. You may say it feels like a meat market, but I prefer to think of it as just a nice place to swim, eat, and mate. Mating, after all, is a significant part of my job, and I excel at it. I hope that doesn’t make you uncomfortable. I do have some outstanding characteristics, and there’s nothing like ultraviolet lights to show them off—my circulatory system, perhaps, or my neurological network. I glow in the dark; we can discuss that further when we’re more familiar.

Danio rerio is my full name, but you can just call me Danio. I’m a zebrafish. And I may be saving your life. In fact, I may be helping you figure out how your life develops in the first place. But that’s you. First, a little more about me. Yes, since you ask, I do always dress this way; I’m only an inch long, so I think the stripes make me look sleeker. And black and white together is, of course, a classic.

Compare me to him over there, in the shabby white fur, muttering to himself. That’s Ratty—Rattus norvegicus. His eyes are red because he’s albino, but also because he’s been drinking himself to sleep since the 1980s, when scientists began doing their first research with zebrafish. Since then I’ve made serious inroads into his territory in the lab. And don’t even ask about that dude asleep with his face on the table in the corner. That’s Cavy—Cavia porcellus. Guinea pig. He’s felt forgotten for decades.

But if you really want to understand what makes me so satisfying as a research subject—a “model system,” as scientists like to call me—visit the labs of some of the Duke scientists doing groundbreaking research using—no, I’m embarrassed. Okay—using me.

Start with cell biology professor Nicholas Katsanis, a geneticist whose research lab uses zebrafish to study genetic diseases in humans. He’s addressing not only which genes cause which diseases in a long-term research-based approach, but also trying to figure out, with the clock ticking, how to address specific genetic abnormalities in newborn babies.

Katsanis says studying genetic disease once it’s either killed a patient or been fully expressed is like archaeology: “We go in after the massacre has occurred, we dust off relics, we can figure out how many soldiers were there,” what weapons they used, and so forth. Which is great—in past tense. “But you cannot use this information in a reliable fashion to predict where the next battle is going to happen or try to prevent it.”

“And I think that’s what it’s all about,” adds Katsanis—actually preventing or curing disease. In the archaeological model, if a child is born with a genetic disease, by the time it’s understood, it’s too late to do much.

“As a community, we are engaged in employing genomics as a brute-force tool,” Katsanis says. “There are three things. Diagnosis? We’re doing okay. But prognosis? And proactive treatment? We really suck at it. I’d like to begin thinking about genetics not in an absolute deterministic terminal disease, but as a manageable set of diseases. How can we use genomics in a cost-efficient manner to improve predictive power? And help inform the decisions made by physicians and their patients?”

This is where I come in. Read this part. It’s about me: Danio.

Most genetic modeling using mammals takes an enormous amount of time to introduce genetic changes and allow them to grow embryonically. A rat or mouse, for example, takes three weeks to gestate maybe a dozen babies, called pups. Plus, of course, you can’t see what’s happening until the pups are born, unless you kill the mother to remove and examine the pups.

Too long for Katsanis, who focuses on babies. “Can we improve the guesses we take as to what’s wrong with these babies, these young kids?” he asks. Can we shorten the diagnostic time and maybe get a chance for treatment?

“If your nervous system is already saturated with some toxin, I can’t fix it anymore. But if we can understand early enough what the toxin is in the first three months, how it gets into your cells, maybe—just maybe—we can do something.”

So enter the zebrafish. The zebrafish allows Katsanis to rapidly address the cause and the treatment of that disease. It shares 70 percent of its genome and its anatomy with humans. It has bones, neurons, eyes, muscle fibers, organs. It shares our fundamental biochemical mechanisms of development from zygote to fully grown animal.

Zebrafish are more than just cheap and easy to manage in large numbers. They're also, at least early in their development, transparent.

“So if I have a child who has loss of light-sensitive cells at the back of her eye,” says Katsanis, “I can actually copy that defect in the developing light-sensitive cells of a zebrafish.” And he can do that cheaply, quickly, and in an adequate number of trials to demonstrate that whatever conclusion the various trials suggest is correct. “Or imagine we have a child who has deteriorating muscles. You sequence the genome, and you end up with twenty-seven candidate genes. Your next question is, which of these genes contribute to that child’s phenotype? Which mutations? And how?” If you used rats or guinea pigs for each of those genes, the experiment would take several years— far too long to help that child—and cost a couple hundred thousand dollars.

But using zebrafish, “we can actually model all these genes in four to six weeks,” says Katsanis. And for a fraction of the cost. Each of these gene candidates can be turned off through the injection of small molecules, known as morpholinos, into an egg. The short embryonic time—two to three days—allows him to examine the effect of inactive genes egg by egg within just a few days. And he can, very directly, watch it develop. Zebrafish are more than just cheap (it costs about 20 cents per day to maintain a mouse; you can maintain about a hundred zebrafish for the same cost) and easy to manage in large numbers. They’re also, at least early in their development, transparent. You can put embryos under the microscopes that feed the big screens in Katsanis’ laboratory, and each looks like exactly what it is—a big egg, with no shell to impede your vision.

Postdoctorate fellow Christelle Golzio gives a tour of the fish rooms. Each shoebox-sized tank can hold twenty fish or more, stacked six-high on long rows of racks. Labels show subjects’ genetic strains and particular experiments. Golzio estimates 10,000 tanks or so, easily a couple hundred thousand zebrafish in this facility—on one floor of one building, and it’s only one of several zebrafish facilities, large and small, at Duke. And if you need more zebrafish? Easy. “We mate them and harvest the eggs,” Golzio says. Which means you put some male and female zebrafish in a tank, leave them overnight, and gather the hundred or more eggs in the morning.

Golzio puts a petri dish with a few eggs on the platform of a fluorescence dissection microscope. The screen shows lovely bluish spheres floating peacefully, with easily identifiable yolks in the middle. A dish from the previous day shows eggs already twitching with life, embryos folded around yolks. They’re still perfectly clear: You can see every aspect of development by doing nothing more complex than looking at a screen. “At two days you can see the blood flowing through the vessels,” she says.

The zebrafish lab population explosion began in the 1970s, when University of Oregon molecular biologist George Streisinger, looking for an inexpensive and effective model for his research, glanced at the fish tank he kept as a hobby. Zebrafish were cheap. They were easy to breed. They lived in murky water in the wild—they were originally isolated from the extremely polluted Ganges River—so they were hardy. They laid and fertilized their eggs wantonly, and the developing embryos were transparent.

He began using the fish in his research and spreading the news of the effective new model. ZFIN (the Zebrafish Information Network), based at the University of Oregon, helped spread word of the model. And as genetic sequencing progressed, the entire zebrafish genome was sequenced. Meaning scientists can, as Katsanis does, choose specific genes, cause specific mutations, and watch the results develop in real time. Katsanis has done this and made a difference in the lives of children. He has letters from grateful families to prove it.

You see? I’m not just a fish; I’m a hero. By 1993 scientists had already held a Zebrafish Meeting at the Cold Spring Harbor lab and decided to rename me: I was called Brachydanio rerio before that; you can read about that in The Zebrafish Book, and yes, you may say it: Book ’em, Danio. Everyone likes to say that.

Back to me. By 1996, the journal Development dedicated an entire issue to me, and in 1998 scientists already had 1,913 publications of research using me as a subject. Last year? Try 17,396, according to ZFIN. 

David Tobin, an assistant professor of molecular genetics and microbiology at Duke, uses zebrafish to study tuberculosis. Zebrafish obviously lack lungs, so Mycobacterium tuberculosis, the bad guy that makes humans sick, doesn’t trouble zebrafish. But, “they do get infected by a natural pathogen that’s its closest relative, called Mycobacterium marinum,” Tobin says. “There are a lot of conservations” between both the two bacteria and the zebrafish and human response to them—places, that is, where the organisms function in the same way. The same kind of immune system cells respond by making the same kind of granuloma to isolate the pathogens. “And the great thing about the zebrafish,” Tobin says, “is, because the larvae are optically transparent, you can visualize the whole process of infection in real time under a microscope, and do long time lapses and look at interactions of specific immune cells with the invading bacteria.

“So that’s really fun.”

The other good thing about zebrafish is what Katsanis especially likes about them: “They’re genetically tractable.” Scientists can go in and affect one or another specific gene, but more important, they can use the fish as a sort of cannon fodder. “About a third of the world has been exposed to tuberculosis,” Tobin says. Yet only 10 percent of that population develops active disease. Environmental and nutritional issues certainly affect susceptibility, “but it’s clear there’s a strong genetic component” to who does and doesn’t get TB. To check that out, you just get tons of zebrafish and tons of Mycobacterium marinum. The zebrafish model allows Tobin to do “forward genetic screens, in which you make no assumptions about what you will find, and you randomly mutagenize the zebrafish.” That is, his lab randomly introduces mutations and other effects and sees what makes fish more or less susceptible to disease—doing so by using fluorescently labeled bacteria and seeing what they see.

Yep—fluorescently labeled. They make bacteria that fluoresce one color and infect embryos that fluoresce in a different color, and then where the bacteria flourish, that part of the fish identifies the site—and allows one to watch the infection. Under the microscope. While it’s alive. And transparent. Katsanis’ lab does the same—introducing fluorescent labeling to elements of the zebrafish, which then makes certain parts of the fish glow.

The heart—at least in zebrafish— can repair itself. This is changing the way medical scientists pursue heart regeneration.

(This glowing, by the way, got started in 1999, when scientists injected eggs with green fluorescent proteins from jellyfish. The Glofish you can buy for your home aquarium are just zebrafish bred to have a lot of that stuff.) Differently prepared markers can illuminate neurons, the circulatory system, bones. Even heart tissue. Heart tissue? According to Ken Poss, professor of cell biology at Duke and Early Career Scientist at the Howard Hughes Medical Institute, zebrafish have yet another capacity that renders them amazingly useful as experimental models. “They do something better than we can: They regenerate tissue.” Poss’ lab studies heart regeneration. Around a million Americans per year have heart attacks, each attack causing the death of up to a billion heart cells. The body responds by quick scarring but cannot generate more heart tissue, and for decades people have believed that heart tissue is a one-shot deal: You get what you were born with, and every time you lose some, it never comes back.

“Zebrafish are the best laboratory system we know so far for regenerating tissue,” Poss says. “If you remove a quarter of a zebrafish heart, or genetically kill two-thirds of their heart cells, they’re lethargic. They don’t perform well on a swim test,” at first. But, in non-human fashion, their heart cells come back?

Seriously. It turns out I can regenerate heart tissue, and instead of a Nobel Prize they give me—I’m a fish!—swim tests. I try to be a good sport.

“Zebrafish can recover, and reverse all these symptoms,” says Poss. “We want to know how they do that.” So amid all the excitement about stem cells and how to teach them to grow into whatever we want them to, his lab has been seeking the source of this new zebrafish heart tissue. Poss’ lab uses the same combination of see-through embryos and fluorescent tagging the other labs use. “We permanently tag certain cell types and follow their progeny,” he says, which led to an astonishing conclusion.

“We thought stem cells would be involved,” he says. “But [zebrafish] have cardiac muscle cells that can be stimulated to divide.” The heart—at least in zebrafish—can repair itself. This is changing the way medical scientists pursue heart regeneration. “Fifteen years ago, there wasn’t the idea or optimism that this could even happen,” he says. “Most thought the heart was an organ that was static.”

Now, he says, human heart regeneration is "something we can almost grasp."

And the introduction of more and more of those fluorescent proteins make the fish more and more useful. “You can get virtually any tissue you want to glow: blood vessels, the heart, the brain, skeletal muscle. We can get maybe fifty different colors in one animal.” You can even tag particular compounds.

“You can, with zebrafish, visualize brain activity. You introduce a tag that will activate when a neuron fires, and you can visualize learning paradigms.” Then you put the zebrafish in a petri dish on a microscope, and you watch it think.

The zebrafish shares 70 percent of its genome and its anatomy with humans. It has bones, neurons, eyes, muscle fibers, organs.

In another Duke lab works Elwood Linney, a professor of molecular genetics and microbiology who pioneered zebrafish research at Duke a decade or so ago. Linney, whose undergraduate degree was in engineering, likes to figure out how things work, and he worked with stem cells in mouse embryos.

“I wanted to work with an embryonic system, where we could watch embryos develop in real time,” he recalls. “I keep on remembering: I’m a visual person.” He switched to zebrafish and hasn’t looked back. “Let’s say you have a gene expressed only in the heart. You identify its DNA elements, clone that, and you couple that to a gene that connects to a fluorescent protein. In the next generation, you get embryos that are expressing that fluorescence in the heart.”

He marvels at Tobin’s research on tuberculosis, where Tobin uses fluorescence to follow infection: “You watch the macrophages come along. You see it happen. Some clever people are making fluorescent probes that are sensitive to calcium,” which is transmitted during the electrical signals causing heartbeats. “You can look at the heart, and the heart is flashing." And the zebrafish heart turns out to beat a lot more like a human's than a mouse's does, so we can learn not just about regeneration, but also about our own heart function through zebrafish hearts.

For his own research, Linney takes Poss’ work even a step closer to the beginning. Every organism, from a single-celled bacterium to a human being starts out as one cell. “You get us from a single cell,” he says. Linney studies how cells know which developmental route to take and when to do so. He’s spent his career studying how the nutrient called retinoic acid, related to Vitamin A, affects the process. Scientists have long known that it plays a role in turning genes on, controlling development, and that its presence at the wrong time can cause catastrophe: Years ago pregnant women using acne creams containing related chemicals saw terrible birth defects.

Current hypotheses suggest that the molecules that allow retinoic acid to turn genes on can, in the absence of retinoic acid, keep genes off at times when they should be off. “There are mechanisms in the fertilized egg that are restricting certain genes and allowing other genes,” says Linney. “We think one of the restrictive mechanisms involves the receptor that retinoic acid binds to.” Linney’s environmental work involves a class of chemicals called triazoles, found in pesticides, body-building compounds, and acne pharmaceuticals. “We think all of these can affect the set of genes that control the distribution of retinoic acid in the embryo, can inhibit the genes that metabolize retinoic acid in the embryo. If the enzymes are inhibited, it’s almost like pouring on extra retinoic acid at the wrong time and mucking things up. We’re using zebrafish as a model for that trying to see which set of these chemicals might be more dangerous for humans to use than others.” He’s also experimenting on human stem cells to examine whether the manner in which retinoic acid receptors restrict gene expression in embryos may also hold in human stem cells.

And who’s helping him do it? Me, that’s who. Danio Rerio. The humble zebrafish, taking over the lab once I got my chance.

All the scientists are careful to note that the zebrafish is hardly a panacea, and that though it’s making a great contribution, mice, rats, nematodes, and other experimental systems are in no danger of being shoved aside. “No one in the zebrafish field would want to give the impression that it’s taking over,” says Poss. “Any model system works: Mice are well established; the tools are there.” Zebrafish are not even particularly personable. “I like my work,” Katsanis says. “I also like sushi. What can I tell you?”

Okay hold on. After all I've done for these scientists...

But it’s a fish that makes science move faster and better, and Duke has jumped on the opportunity, Poss says. “Our hope is to use the system in a way that provides the advantages that are lacking in mice: speed, throughput [high-volume, rapid experimentation], and visualizing animals early in development.” He’s enjoyed watching the zebrafish community spread over the past ten years, since Elwood Linney was at Duke essentially on his own. “The school has been especially supportive. It’s inexpensive to work with zebrafish compared to mice, but to build the facility? The equipment? That’s a big investment, and not every place will make that investment.”

It’s a big investment in a small fish—a small fish that deserves credit for helping to unlock some very big secrets.

Okay, I can work with that. But I’ve actually got to leave you now. Both Ratty and Cavy are awake, and neither of them looks cheerful. I wonder if you can tag your genes for invisibility?


  • Scott Huler is the senior writer at Duke Magazine.