When one Carolina researcher looks at a lab mouse, he sees what most researchers see: a crucial player in the battle against human diseases.

But he also sees a major problem.

For Fernando Pardo-Manuel de Villena, the problem lies within the lab mouse’s DNA. Pardo-Manuel de Villena and a team of researchers at Carolina and the Jackson Laboratory in Bar Harbor, Maine, took a closer look at the genes of mice studied in labs. And their findings were surprising.

Turns out that the most common types of lab mice studied today represent only a fraction of the genetic diversity found in mouse populations.

So what’s wrong with that?

The most common diseases, such as heart disease and obesity, are also the most complex, affecting multiple genes at different locations in our DNA. Understanding the genetics of these diseases requires the ability to study an organism exhibiting variation in DNA sequences between different strains, or families, of lab mice, Pardo-Manuel de Villena says. Different strains of the commonly used lab mice simply don’t have this kind of genetic diversity, or genetic variation.

Mice and men

The genetic history of lab mice is intertwined with a social history of humans. Mice were first domesticated as pets in China about two thousand years ago. When Chinese mice were brought to Europe in the nineteenth century, a pet mouse craze exploded. “Almost all wealthy Western European women would have mice as pets — more than cats or dogs,” Pardo-Manuel de Villena says.

Breeders in Asia and Europe bred mice to produce attractive pets called fancy mice. Breeders could produce offspring with a range of sizes and coat colors, including blue and albino, and with characteristics that were more attractive to buyers and less useful for the mouse as a species. A popular type of “waltzing” mouse, which looked like it danced in circles, actually suffered from a kind of seizure, Pardo-Manuel de Villena says. Pet mice became bigger, fatter, and more docile than their wild cousins.

Fancy mice spread from Europe to the United States and, in the early twentieth century, into the hands of Harvard scientists. To them, fancy mice seemed ready-made for research. Thousands of generations of selection for specific characteristics had resulted in mice that were nearly identical. Researchers, whose work relies on carefully controlled laboratory conditions and experimental replication, could produce genetically identical mice by mating brothers and sisters from these fancy mice over several generations.

Like the Europeans who selected their pet mice by color, scientists choose to study certain mice based on their susceptibility to cancer and other diseases. Like breeders, geneticists manipulate mouse DNA, albeit more directly, using techniques to inactivate one or more genes (see “A Life at the Bench”).

But scientists conduct experiments almost exclusively on a few fancy-mouse strains, Pardo-Manuel de Villena says. The centuries of inbreeding almost certainly resulted in strains of lab mice with little genetic diversity. But how little?

Mapping mouse genes

To measure the genetic variation in populations of lab mice, the team compared the DNA strains (called classical inbred strains) to more natural mouse strains (called wild-derived strains).

Using data from the National Institute of Environmental Health Sciences, the researchers looked for differences among four wild-derived and eleven classical inbred strains at the smallest subunit of DNA — the base-pair. If you think of DNA as a large spiral staircase, the base-pair would be a single step. The number of variations among the types of steps indicates how different each mouse strain is from another.

The team found a surprisingly high level of variation — about forty million among the fifteen strains. In other words, the genetic variation among the three sub-species of mice may be analogous to the genetic variation between humans and chimpanzees. “To us, mice don’t look that different,” Pardo-Manuel de Villena says. “But they are different.”

Pardo-Manuel de Villena’s team found almost five times more variation than a competing studying using the same data. That’s because his team analyzed the DNA data in a different way, he says, sequencing multiple mouse strains at many locations in their DNA. The competing study used an old model that is disproved by the more recent and comprehensive data, he says.

But of the forty million variations found, most were not present in the strains of mice that are commonly used in labs. Wild-derived mice contained ten times more genetic variation than classic lab mice, Pardo-Manuel de Villena says, and the genetic disparity between strains implies that existing genetic research involving lab mice has limited applicability.

Not all research on lab mice is problematic, though. Classic lab mice are still useful for studying diseases that affect one gene at a time, such as cystic fibrosis. But the findings point out the need for “a new population of mice that has more diversity,” Pardo-Manuel de Villena says.

The “model” model

Pardo-Manuel de Villena and scientists at several other universities are working to fill that need, creating new strains of lab mice specialized for complex disease research — strains that maximize the amount of diversity.

The group, called the Complex Trait Consortium, mixed classical and wild mouse strains together and bred a new model called the Collaborative Cross. Each progeny has a “checkerboard of variation,” says David Threadgill, a Carolina researcher involved with the Collaborative Cross. He says the group plans to produce a thousand highly diverse, highly randomized mouse strains. In two years they will have about half of the strains developed, he says, but some strains are already under evaluation in cancer-related studies.

In time Threadgill hopes the new mouse strains become a genetic reference population, so they can be used as a common denominator to support data integration from all types of biomedical research. The Complex Trait Consortium has a long way to go — 95 percent of all mouse research has involved descendants of the European fancy mice.

But Pardo-Manuel de Villena says that scientists are starting to warm up to the idea of a new type of mouse model for research.

“People have come to realize the severe limitations of classical mouse strains for genetic studies,” he says. “And they’re beginning to see the very, very large pool of genetic variation in the mouse that can be tapped by using wild-derived strains.”

Sarah Whitmarsh was a student who formerly contributed to Endeavors.

Fernando Pardo-Manuel de Villena is an associate professor in the Department of Genetics. His study was published in the online version of Nature Genetics in July 2007. The National Institute of General Medical Sciences, part of the National Institutes of Health, funded his research.

David Threadgill is an associate professor in the Department of Genetics. The Collaborative Cross receives funding from the Ellison Medical Foundation and from the National Institutes of Health.