Hertz Fellow Michael Baym Tipping the Scale against Our War with Lethal Antibiotic Resistance

October 2, 2016

A powerful tool for observing bacteria as they mutate and develop resistance to antibiotics has been developed by a team from Harvard Medical School and Technion-Israel Institute of Technology, including Hertz Fellow Michael Baym, a Harvard postdoc.

The device is a 4-foot by 2-foot petri dish with a black, inked background that makes the bacteria more visible for time-lapse photography. The video, "The Evolution of Bacteria on a 'Mega-Plate' Petri Dish", clearly shows the spread of two colonies of bacteria across zones of increasing antibiotic concentration over a two week period.

Beyond the dramatic demonstration of the development of antibiotic resistance, which may be useful for increasing public awareness, Baym says the setup can contribute scientifically to studies such as comparing and contrasting the rates of growth of the two colonies of bacteria; following specific lines of mutant bacteria as they compete with one another; and extracting samples of successful and failed bacteria for analysis of the reasons for their performance. In the longer run, he says, he and the research team he is part of hope such research will help them gain new insights that let them stay one step ahead of bacterial evolution and reduce if not prevent the development of antibiotic resistance. An account of the development was published in the September 9 issue of Science magazine by Baym and six other authors. The work was done in the laboratory of Roy Kishony, a systems biology professor at Technion-Israel Institute of Technology and Harvard Medical School. A third member of the team was Tami Lieberman, then a graduate student at Harvard and now a postdoc at MIT.

The researchers call the petri dish MEGA-plate, for Microbial Evolution and Growth Arena plate. It was inspired by a video featuring a stylized petri dish advertising the 2011 movie Contagion about a fictional epidemic caused by the spread of a deadly virus. Kishony and Liberman saw the scientific and instructional value of watching evolution take place in a giant petri dish and brought Baym as part of the design and development effort. Baym refined a prototype through a series of steps into the device described in the Science article.

On a personal basis, Baym had not always been interested in the experimental side of things. He studied applied math in graduate school at MIT starting in 2003, concentrating on computational issues in biology including mathematical and algorithmic problems of very large scale genomics and epidemiological modeling. He was turned down for a Hertz Fellowship twice before he finally won one in 2004. The experience reinforced a lesson he had learned from a decade of studying and teaching the Korean style of karate, Tae Kwon Do, in which he has a fourth degree black belt. “When you get knocked down, you get back up,” he says.

With the award of the Hertz Fellowship, he had the financial independence to try branching into experimental science. Far more important, he says, a 2008 Hertz Foundation retreat at Woods Hole brought him face to face with other Hertz Fellows and Hertz Foundation officers, highly creative people like Philip Eckhoff and Lowell Wood who take an interest in many fields of science and stimulate others to do the same. “Just meeting other Fellows gave me the sense that maybe it was possible for me as a computations guy to learn how to do experiments,” he said. “I was very inspired by them.” It was a conversation with Hertz Fellow Jeff Gore at that retreat, in fact, which started him thinking about antibiotic resistance and connected him with Roy Kishony’s laboratory, where he has worked since 2009.

Baym says that when he first saw the now viral video, he was “shocked.” He guesses that he watched it “25 times in a row” and now has seen it “something like 1,000 times.” It has been seen by others an estimated 23 million times. In the video, colonies of non-pathogenic E. coli bacteria are started at opposite ends of the MEGA-plate. They spread readily across a zone with no antibiotic to the start of a zone with just enough of the antibiotic trimethoprim to stop them. Trimethoprim is commonly used in combating bladder infections.

After a few hours, some of the E. coli experience a spontaneous mutation that allows them to grow in the low-dose trimethoprim zone. This happens at several points along the boundary between the two zones, and the low-dose zone is soon overrun with competing mutant strains of E. coli. Baym says that about one in a thousand of the bacteria experience the mutations.

The next zone has a level of trimethoprim 10 times higher than the original low-dose zone. It stops the E. coli for a time, and then the bacteria break and soon flood the zone. The same thing happens with trimethoprim levels 100 times and finally 1,000 times higher. At that point, about 11 days after the original colonies were planted, the entire plate is overrun.

Baym, Kishony and Lieberman have made it clear in various interviews that they are strongly in favor of more public attention to the resistance problem. There seems to be a disconnect between vague, general knowledge that there is such a thing as bacterial resistance and the down-to-earth fact that people have to take steps to prevent it.

The problem is large and growing. In 2013, the Centers for Disease Control reported that more than 23,000 people in the U.S. alone die each year from direct infection by antibiotic resistant bacteria, and far more than that die of other conditions made worse by resistant infection.

The problem is hardly new. In doing research for a survey article, Baym found a 1941 reference to the first experimental evolution of resistance in the laboratory, from successive exposures of staph aureus to higher doses of penicillin. Scientists have been studying the problem for three-quarters of a century.

MEGA-plate is potentially powerful from “advocacy and public policy perspectives,” Baym says. “One of the reasons antibiotic resistance is such a growing problem is that a lot of things we do are making it worse. We really need to stop using antibiotics when we don’t need to…in agriculture, when people have colds, things like that.

“I hope the attention this is getting will help increase awareness and start to inspire the kinds of collective and governmental action needed for this to take place on a population level.” As dramatic as the MEGA-plate video is, Baym thinks it telling that resistance isn't even more widespread, "If antibiotic resistance is this easy to develop, why is it that antibiotics work at all?", he wonders. “This sort of implies that there are other dynamics in nature that keep resistance in check.”

He believes antibiotic resistance has been “treated almost as a black box. We come up with a new drug, resistance inevitably evolves, and we have to find a new drug to replace the first one.” MEGA-plate should be helpful in further scientific studies. By analyzing the genetics and performance of mutant strains of bacteria using MEGA-plate, the Kishony team is hoping to be able to discover evolutionary factors that might be exploited in a more generic way, something like “playing one move ahead in chess,” as Baym puts it.

“We can be sure that bacteria are going to evolve in response to pressures that we put on therm. If we can predict ways they are going to evolve and maybe control that, we have some chance of tipping the scale against them.”

Our War with Lethal Antibiotic Resistance

The MEGA-plate at the end of a 12-day symmetric trimethoprim resistance evolution experiment. E. coli appear as white on the black background. The 182 sampled points of clones are indicated by circles, colored by their measured MIC. Lines indicate video-imputed ancestry.