Kellie Woll administered a few squirts of clear liquid to a dish full of wriggling tadpoles and within minutes, the creatures became completely still.
A few minutes after that, they started to move again. No surprise, as the liquid contained propofol, a widely used anesthetic.
With it and most other anesthetics, however, there is not much difference between the amount needed to put someone to sleep and the amount that will knock one out permanently.
Woll works in the University of Pennsylvania lab of Roderic G. Eckenhoff, who is on a long-term quest for better alternatives.
The researchers are interested both in preventing death from anesthesia - which today occurs in fewer than one in 100,000 cases - and addressing more subtle side effects. Chief among them is the possibility, still unproved, that multiple exposures to certain anesthetics can contribute to cognitive problems in the elderly and the very young.
And tadpoles, of all things, could be part of the solution, said Eckenhoff, a professor at Penn's Perelman School of Medicine.
In February, he and colleagues reported the discovery of two compounds that were effective anesthetics in tadpoles and in mice.
Most drug candidates that seem to work in animals don't translate to human use. But even if that turns out to be the case here, what's more important, Eckenhoff said, is the new drug-discovery process that he and colleagues developed, given that the drug industry has not come up with any truly new general anesthetics since the 1970s.
The Penn team started by using a robotic system to screen more than 350,000 drug candidates from a library of compounds at the National Institutes of Health. That process eventually led to several dozen promising substances the team tested on tadpoles. Four of those were then tested on mice, and two were especially effective as anesthetics yet had few side effects.
"We don't know if these two molecules are ultimately going to make it into people," Eckenhoff said. "But the process is very likely to produce molecules that we can put into people."
Douglas E. Raines, a Harvard Medical School professor who was not involved in Eckenhoff's research, agreed.
"The approach is almost as important as the actual drugs that he discovered," said Raines, an anesthesiologist at Massachusetts General Hospital. "He's cast a wide net, looking for new and novel compounds that don't look anything like what we have now."
Drug companies have used such robotic methods to search for new drugs in other areas of medicine. Called high-throughput screening, the technique has met with limited success. More commonly, companies tweak existing drugs, and Raines has used that approach to modify one kind of anesthetic to have fewer side effects.
But the high-throughput method is a fast and relatively inexpensive way to evaluate many thousands of compounds, and it had not been used to look for anesthetics, Eckenhoff said.
"We figured it was time to try it," he said.
In brief, the robotic system identified compounds by measuring their ability to bind with a certain kind of protein - one that was known to have an affinity for existing anesthetics.
By itself, a positive result on the robotic screen did not mean a particular substance would work. All it meant was that in one respect, it behaved in a chemically similar manner to known anesthetics.
That is where the tadpoles came in.
The creatures are the size of a fingernail clipping, yet they tend to respond to anesthetics much the way humans do. As vertebrates, they have more in common with people than do some other common lab subjects, such as fruit flies.
And because these tadpoles are translucent, scientists can see their hearts beating even when the rest of their bodies are motionless.
Woll, the graduate student, demonstrated that recently by adding different concentrations of anesthetic to three dishes of simulated pond water, each containing a few dozen tadpoles. Generally, the creatures that received higher doses took longer to wake up.
Woll, who is working toward a doctorate in pharmacology, tapped a pen on the side of the dish that got the highest dose. The tadpoles remained still, their tails extended straight out.
"That's a lack of startle reflex," she said.
In addition to measuring the startle reflex, the scientists also record whether the animals display spontaneous movement.
Future studies in mice - which, as mammals, are even closer relatives of humans - may help the researchers tease out which drug candidates are least likely to have unwanted side effects in people.
Nausea is a fairly common side effect with current anesthetics, as are changes in heart rate and blood pressure, but anesthesiologists have become skilled at monitoring vital signs and responding accordingly. That can be seen in how the rate of death from anesthesia has plunged. Just a few decades ago, it was 2 in 10,000 - more than 20 times the rate today.
Still, it happens, as with comedian Joan Rivers, who died last year after being given anesthetic in an outpatient clinic.
Less clear are the possible links between anesthesia and cognitive problems. Eckenhoff's research has found that elderly patients undergoing multiple courses of general anesthesia may be prone to having abnormal clumps of proteins build up in the brain - a hallmark of neurodegenerative diseases such as Alzheimer's.
Similarly, some studies suggest children under 3 who are anesthetized multiple times have a higher rate of learning disabilities.
But some of that may be caused by the stress of the surgery itself or the underlying condition for which the surgery was performed, he said.
Drug companies have not come up with a truly new general anesthetic since propofol was developed in the 1970s. Erik Gordon, an assistant professor at the University of Michigan business school, said that's because companies are loath to risk research dollars in a field where current products work well enough.
"If an academic lab does come up with something that's a huge game-changer, a drug company will license it," Gordon said.
And if the Penn scientists have their way, they will get there someday, with the help of small creatures swimming in a laboratory dish.