- By Gary Wolf
In 1957, a new medicine appeared on the market. Thalidomide was an effective sedative, but it was also promising as a treatment for pregnant women because it quelled nausea and vomiting. And scientists had great confidence in thalidomide’s safety. It had been tested extensively on mice and found to be nontoxic. It was so harmless, in fact, that no lethal dose could be established.
But thalidomide was withdrawn from the market after only a few years, by which time its name had become a byword for the horrible consequences of placing too much faith in the similarities between mice and men. Although nontoxic in rodents, thalidomide caused human limbs to stop growing prematurely in utero, resulting in the birth of babies with malformed arms and legs. It took decades for the science behind this medical catastrophe to be fully understood: The shape of the thalidomide molecule fits neatly into a stretch of human DNA that controls the formation of new blood vessels during a few short weeks of embryonic development. Before this time, thalidomide has no effect. After these weeks, it is equally safe. During the years of its use, from 5,000 to 10,000 children around the world suffered severe birth defects.
The history of using mice to stand in for humans in medical experiments is replete with failures. The case of thalidomide is only the most notorious. Cancer, cardiovascular disease, diabetes, Down syndrome — mice that express some version of each of these conditions sit at the foundation of multiple lines of research, all of which have had major setbacks. The reason became clear over decades of fervent and often fruitless science: Mouse metabolism is not human metabolism, so mouse-based diabetes and cardiovascular studies may be fatally flawed. Mouse cancers are not human cancers, so oncology models can be misleading. Mouse Down syndrome is not human Down syndrome, so conclusions drawn about the disorder may be wrong.
Barely 10 years ago, somebody who took a quick glance at the history of the mouse in science might reasonably surmise that the story was reaching its end. “There is no question about it: The number of animals used in laboratory experiments is going down,” a Scientific American article concluded in 1997. But this analysis was incorrect. Instead, the opposite happened. In the decade that followed, the lab mouse had a sudden and dramatic resurgence that continues today.
Last year, the Jackson Laboratory, historically the most important supplier of lab animals to science, sold more than 2.7 million specimens, up from 1.9 million a decade earlier. Jackson’s major competitor in the mouse business, an operation known as Charles River Laboratories International, sold more than $660 million worth of rodents and related services in 2008, up from $405 million in 2003.
What happened? The answer is that the mouse changed. Our old laboratory mice had been bred to resemble us in interesting ways, to suffer familiar diseases like diabetes or cancer, or to achieve impressive goals, like extreme longevity. They were model animals in the sense that they were used as substitute people, miniature humans that were both versatile and morally expendable. The new mouse is a model in a different way — not a tiny stand-in for a human but a kind of exemplar. We are using it to explore the limits of biological systems, sending it into a future where flesh is blended with code.
The archetype of the laboratory mouse was invented early in the last century by an undergraduate at Harvard. For three years, starting in 1909, C. C. Little, a former college track star whose father had bred dogs, mated generation after generation of mouse siblings. The word genetics had been coined only a few years earlier, and many researchers were trying to find out whether the laws of inheritance discovered by Gregor Mendel through experiments on peas in the mid-1800s could be replicated and extended with other species. The structure of the DNA molecule wouldn’t be known until 1953, and figuring out how to sequence genes would take still more decades; early geneticists were working in the dark. They inferred things about the genome by mating animals and hybridizing plants, looking for patterns and ratios in the traits that were inherited. To reduce the complexity of their experiments, they often bred close relatives. This allowed them to stabilize a genome, fixing traits through generations. Fruit flies, an early favorite of geneticists, could be inbred easily. But in mammals, each incestuous generation was weaker than the last.
Little’s approach to the problems caused by inbreeding was straightforward: He simply used more mice. The healthiest offspring of each generation were the progenitors of the next; weak individuals were culled. In three years of experiments, Little reared more than 10,000 mice. By 1913 he had a healthy, genetically stable inbred strain. He could then choose a characteristic — pink eyes, say, or a brown coat — and produce an endless supply of animals, each as alike as twins or clones. Some researchers turned to guinea pigs or rats to explore the workings of mammalian genetics. But for Little, mice turned out to be a lucky choice. They were small, docile, and cheap to feed. And for reasons that wouldn’t become clear until much later, they were easier to manipulate genetically than other rodents. But there was one special feature that sealed the success of the mouse over all its competitors in labs around the world: The mouse got cancer.
Wired
