I’ve killed birds by suctioning their brains, and I’ve sacrificed the lives of tiny kittens. I’ve withheld water from a live macaque. But in all the time I spent in graduate school for neurobiology, the most diabolical procedure that I performed in the name of scientific research—the one that haunts me still—was my very first: the perfusion of a mouse.
That begins with the animal etherized upon a lab bench, its paws spread wide and secured with tape. I’d snip the skin of its abdomen with a pair of nail scissors, and separate the muscle to expose its heart, smaller than a hazelnut and still beating. The mouse’s own circulatory system would become my instrument, pumping a preservative throughout the mouse’s veins and arteries, so as to pickle every cell inside its body. First I’d poke a hole in the right atrium of the heart, allowing all the blood that’s flowing in to drain into the body cavity. Then I’d slide my syringe into the left ventricle and start feeding it with paraformaldehyde—only so much as the organ could press out with its diminishing contractions. A few minutes later, all the mouse’s vessels would be flushed, made pallid by the chemical, and repurposed for my use. The animal, split open on the table, would have mummified itself before my eyes.
I don’t believe the mice suffered all that much during this process—I’d pinch their feet as I anesthetized them to make sure they were really out, and the procedure had been approved by a standard institutional review committee at my school. But in truth I’d never worked with mice before and had no way to intuit their well-being. (It’s much easier to see pain in the face of a monkey or a cat.) At any rate, it was clear enough that these rodents were disposable: If one perfusion didn’t go so well, I’d simply reach into the cage and grab another mouse. It seemed as though the rules for working with these animals were different from the ones for other species. Bigger, more likable animals were sacrificed for research. Mice were used.
I only learned much later that this disparity has been ensconced in U.S. law for half a century. The Animal Welfare Act, signed by President Lyndon Johnson in August 1966, enforces an apartheid system of concern for animal suffering. This has been the case for most of the law’s history, though in its original formulation, it was intended to protect all warm-blooded animals in the lab. (It was also meant to address a national panic over dognapping.) “Science and research do not compel us to tolerate the kind of inhumanity which has been involved in the … careless and callous handling of animals in some of our laboratories,” Johnson said that summer, at a signing ceremony that was strategically small and out of sight. But within a few years, the U.S. Department of Agriculture—the agency responsibly for implementing the new law—had decided that the rats, mice, and birds in science labs were simply too numerous to be protected. “The math just did not allow it,” a former department official told me several years ago, for another article in Slate. “We couldn’t use the National Guard to make all of these inspections. We didn’t have the force.” On the basis of that calculation, these species were formally excluded.
As we approach the 50th anniversary of the Animal Welfare Act, it’s worth reviewing the legacy of this decision, in both practical and symbolic terms. Rats and mice remain trapped in a special, unprotected category, although they compose a large majority of research animals around the world. Efforts to refine experiments and make them more humane has led to broad reductions in the use of many different species: Numbers from the USDA demonstrate that experiments on protected animals—dogs, cats, and monkeys; rabbits, hamsters, and guinea pigs; and so forth—have been dropping over time and are at a record low. Still, figures on the use of rodents suggest a rapid upward trend. In European labs, rats and mice together represent about four-fifths of experimental vertebrates (a category that includes mammals, fish, birds, reptiles, and amphibians), and their global use now results in more than 100,000 published papers every year.
Rats and mice were always popular, but today’s focus on genetic engineering—creating study animals with specific genes knocked out or inserted—has redoubled the rodent cull. The standard method for producing a genetically modified line of research mice requires careful interbreeding over several generations. All those extra mice produced along the way are put to death. Data from the Netherlands suggest this latter group of animals, those “bred but not used,” are just as numerous as the ones that die in research tests. And in the U.S., at least, there hasn’t been much of an effort to figure out exactly how the state of being genetically modified might affect a mouse’s welfare. Should the modification of a mouse’s genome—in a way that gives it something analogous to Alzheimer’s disease, for example, or makes it prone to getting cancer—be counted as a source of suffering? It’s a tricky question, and not so often puzzled out.
Meanwhile, there’s evidence—more objective than my memories of grad school—to suggest that rats and mice are not being given the consideration they deserve. One survey of the research literature, published in 2009 by a group led by Claire Richardson at Newcastle University, looked at studies involving surgeries performed on rats and mice and counted how often papers noted the use of full-fledged pain management. Just 20 percent of these studies reported giving rodents systemic painkillers (and that’s even considering that the survey found signs of improvement in recent years). A recent update focused on one set of studies where scientists would open up a mouse’s abdomen, lift its liver and tie off the bile duct. Out of 119 papers describing this procedure, the authors found that all of 28 percent mentioned using anesthetic with pain-killing properties.
What about other species? Richardson and her colleagues discovered that scientists were far more likely to describe using full-body analgesics while experimenting on rabbits, pigs, sheep, dogs, or monkeys. For those studies, the rate was about three times higher, at 63 percent.
It’s not quite right to blame this double standard on the Animal Welfare Act’s exclusions. While rats and mice aren’t covered by the AWA, their treatment is still subject to guidelines from the National Institutes of Health, which exerts tremendous influence on institutions via the tens of billions of dollars it provides for biomedical research. According to Brianna Gaskill, assistant professor of animal welfare at Purdue University, the rules end up looking pretty similar.
Gaskill sees a more fundamental problem, though. “It’s something that we’ve realized in the last six years,” she says: “We are really bad at determining pain in rodents.” That’s the feeling that I had in grad school, too—next to dogs or cats or monkeys, it’s very hard to figure out when a mouse is suffering and to what extent. This opacity could be a result of the mouse’s ecological niche. A vulnerable prey species might be better off concealing its distress so as to hide its vulnerability. Recent work by pain researcher Jeffrey Mogil has helped reveal some subtle signals. Now we know that mice will show their discomfort by making tiny grimaces. Mogil has also demonstrated, on the basis of these grimaces, that the painkillers most often used in experiments are not effective at reducing a mouse’s discomfort unless they’re used at up to four times what is currently the recommended, standard dose.
Even as lab rodents multiply and spread around the world, the science of their welfare remains in an embryonic state. We don’t even know how best to lodge the mice and rats in cages. Until very recently, for example, mice were kept at temperatures that were cold enough to stress them out and raise their metabolic rates. (This isn’t simply a moral concern: Stressed-out mice can bias the results of experiments.) For her dissertation research, Gaskill found that the obvious solution to this problem—turning up the thermostat—wasn’t all that helpful. “The mice want different temperatures for different parts of the day,” she says. “There is no way that humans can pick a single, perfect temperature for mice.”
So she devised a better way to make the mice more comfortable. Instead of turning up the heat, she put crinkled paper in their cages and let them use the stuff to build a nest. That made them healthier on several measures, including higher rates of reproduction. Charles River Laboratories, a global breeder and distributor of research animals, has since picked up on her idea.
It may seem like a no-brainer that an “enriched” environment including nesting materials would improve the lot of laboratory rodents, but other natural fixes may have steered us in the wrong direction. For example, the NIH recently changed its guidelines to recommend bigger cages for breeding female mice. But Gaskill’s research found this had no effect; the mice didn’t seem any better off in deluxe accommodations. Other work has shown that living in a smaller space even seems to improve the immune function of captive mice and lead to less aggression among cage mates.
It would be nice if we could figure out exactly how to make the mice as comfortable as possible, given their place at the bottom of the lab-animal pyramid. But Gaskill says there simply hadn’t been enough rigorous research to answer fundamental questions, in part because there hasn’t been much money for research on on the topic. Her own dissertation research was only funded by a fellowship from a U.K.-based organization. (In general, European countries have been far more aggressive than the U.S. at regulating the use of laboratory animals. The E.U. covers every kind of vertebrate, along with squid, octopus, and their fellow cephalopods. Crabs, shrimp, and lobsters were even considered, but the evidence in their favor was deemed lacking.) U.S.-based grants for such studies, from various laboratory-animal science associations, top out at a rather paltry $50,000, Gaskill says. Now she claims to be one of just two professors in the U.S. who specializes in lab-animal welfare, and she may soon be forced to switch her focus to farm animals, where grants are more forthcoming.
Even if there were more money for researchers such as Gaskill—and even if these researchers were able to devise new and better ways to study rats and mice—it would still be hard to keep track of real improvements to their welfare. Thanks to mice and rats’ exclusion from the AWA, no one keeps official tallies of their use. They’re simply missing from the annual USDA reports. So even the most basic welfare question can’t be answered with authority. Is the number of rats and mice used in labs going up or down? We can only make an educated guess. Assessing how many have proper housing or how many are being adequately anesthetized is difficult without such standard baselines.
The first time that I perfused a mouse, I felt rattled for the rest of the day. I’d opened up the animal, turned its chest inside-out, and used its beating heart to my advantage. The process left me in a state of queasy wonder, sadness, and confusion at how utterly one can utilize a living thing. Naturally this feeling weakened over time, draining out a little more with every run-through of the protocol. Familiarity became an analgesic: I went numb.
It’s likely that some 80 million rats and mice are killed in U.S. research labs every year. Even half a century ago ago, their use was so widespread as to paralyze the government: Having just passed the most important animal-welfare legislation in the nation’s history, regulators balked at its enforcement and then modified the mission of the law to eliminate the problem. Nothing much has changed: Lab rats and mice are more common than they’ve ever been, but their suffering remains a willful mystery.