Taking on TB

Walking into Bree Aldridge’s eighth-floor lab at Tufts University School of Medicine in Boston is not for the faint of heart. Step one, empty your pockets—phone, wallet, keys. What goes in doesn’t come out—not even the air, and definitely not the lab notebooks, which is why the windows that look onto the hall are littered with Post-it notes of data points. Next comes the first layer of personal protective gear: respirator, gloves, hair cover, safety glasses. Finally, suit up in a pair of white Tyvek coveralls, booties, a second pair of gloves, and sleeve protectors, with tape securing every opening. A red line on the floor marks the point of no return; step over that, and you’re officially ready to start the workday.

“It takes a couple months to get used to how it is, and then it’s sort of like riding a bike,” Aldridge said. Now a ten-year veteran of the process, she can do it in five minutes flat.

The reason for the precautions is that Aldridge and her eight-person lab study tuberculosis. It’s one of the deadliest diseases in the world, responsible for up to one billion deaths over the last 200 years. Researchers can study it only in the confines of what’s called a Biosafety Level-3 laboratory, the same level of biosafety required for anthrax, SARS, and other lethal pathogens.

Aldridge’s BSL-3 lab was built specifically for studying TB—in fact, she is the Tufts University School of Medicine’s first-ever hire in what the school hopes will become an innovative new TB research center. Such investigation is crucial: Besides being among the world’s most prolific and difficult-to-treat killers, the bacteria that causes tuberculosis, Mycobacterium tuberculosis (M. tuberculosis) can’t be easily studied with traditional microbiology. Even more than a century after its discovery, remarkably little is known about the microbe’s most basic properties, including its behavior, growth, and life cycle.

But Aldridge is no traditional microbiologist—she’s a bioengineer by training, with talents for intense computational modeling and mathematics that have already helped her break at least one new paradigm in the TB field. And now, with an interdisciplinary team of engineers-turned-biologists, and biologists newly trained in heavy computation, she has joined the vanguard in a new era of tuberculosis research, where her contributions could revolutionize how this deadly disease is understood and treated.

Consumption” got its nickname for the way its victims waste away, as if consumed from the inside by their illness. Jane Austen is believed to have died from consumption as is gunslinger Doc Holliday. But TB is no scourge of the past—it is very much alive today, and growing stronger all the time. It recently had the dubious honor of exceeding both HIV/AIDS and malaria as the leading cause of death in the world from a single infectious agent. In 2016, more than 10 million people fell ill with tuberculosis; 1.7 million of them died from it. That’s the equivalent of the entire population of Portugal getting sick every year, and two-and-a-half times the population of Boston dying.

If those numbers seem high, they pale in comparison to the fact that a full quarter of the global population carries the disease in its latent form—that’s more than 1.9 billion carriers—and some 190 million of them are expected to one day get sick.

“TB is among the top infectious diseases affecting human health across the globe,” said Jenifer Jaeger, who directs the Infectious Disease Bureau at the Boston Public Health Commission. “It’s a huge public health issue.”

Most of these cases are in middle- to low-income countries, such as India, Indonesia, and China, but every country is affected, including the U.S. Tuberculosis is even in Boston, where Jaeger’s office helps manage a small, but steady number of active and latent cases every year.

No matter where tuberculosis turns up, Jaeger explained, treating it can seem nearly as onerous as the disease itself. Active cases require taking four or five different drugs for at least six months, often more. For drug-resistant tuberculosis, which is on the rise, doctors must turn to second- and third-line antibiotics, administered for up to two years. It can be a brutal regimen, with possible side effects that include hearing loss, blurred vision, psychiatric disorders, hypothyroidism, and organ damage. Even then, nearly half of these cases will not be cured, according to the Global Alliance for TB Drug Development.

None of that was on Bree Aldridge’s radar during the first chapter of her career. She was busy studying computer engineering and cellular biology in Arizona, then bioengineering for her Ph.D. at MIT, where she built cutting-edge mathematical models of cancer cell death. It wasn’t until 2007, when she was nearly done with her MIT degree and on vacation in India, that she began to think seriously about infectious disease. India is a hot spot for many illnesses, including TB and HIV/AIDS, and as Aldridge walked through the streets of Mumbai, she observed many people coughing and obviously sick. “It made me wonder if maybe I should think outside of cancer,” she recalled.

When she returned to Boston, Aldridge began to explore infectious diseases and quickly found herself fascinated by a puzzle that perplexed tuberculosis researchers. Why were the treatments so onerous? Most of the bacteria died in the first two weeks of treatment, just like any other infection. It was actually similar to cancer that way, she realized. “Both require multiple drugs for extended periods of time, and standard treatments don’t always work,” she said. “It’s because of relapse. It’s because there are some proportion of the target cells don’t get killed by the drug therapy. Why is that?”

The real reason researchers couldn’t perfect the drug treatments, she learned, was because they didn’t know exactly what they were working with. Experiments with mycobacteria, the bacterial group that includes tuberculosis, notoriously produced more varied and fuzzy results than work with better-known organisms, such as Escherichia coli. But nobody could figure out why, because the bacteria were too hard to observe. Mycobacteria grow incredibly slowly—it takes 20 to 22 hours for an M. tuberculosis cell to divide (another mycobacterium, M. leprae, which causes leprosy, takes 14 days). E. coli, by contrast, divide about every 20 minutes. For a long time, researchers couldn’t use traditional microscopes, because the light from the scope would kill the cells before they had a chance to grow—and if that didn’t get them, the prolonged time in a standard culture dish without a steady flux of nutrients and oxygen would.

So for decades, scientists did the next best thing. They studied easier bacteria and extrapolated, as many researchers do with mice when they want to ask questions about humans. “And that was fine; many of the bacteria they studied looked pretty similar,” explained Chris Sassetti, who studies tuberculosis at the University of Massachusetts Medical School. “But what that ignored is that different species of bacteria are as distinct from each other as we are from yeast.”

Things finally began to change by 2008, when Aldridge signed on as a postdoc in the lab of prominent tuberculosis researcher Sarah Fortune at the Harvard T.H. Chan School of Public Health. Aldridge’s own field of bioengineering had actually devised a lot of new methods and technologies that might be able to overcome challenges in studying TB. And that’s when Aldridge realized she could now do something no one ever had before: look at the tuberculosis-causing bacterium as it lives and grows.

The first time I saw them, I thought they were really weird,” Aldridge said. We were sitting in her office on the Tufts’ Health Sciences campus, watching a video she had captured of TB bacteria dividing in culture.

Mycobacteria are named for the study of fungus—mycology—and it suits them. On a time-lapse video, they grow in an eerily slow, staggering motion. They’re shaped like small, irregular rods, sticking close to one another as they pinch themselves off into new segments. The effect is that of slowly unfurling spider legs.

Getting video of these microbes for the first time wasn’t easy. Although new microscopes and recording tools were available, it still took Aldridge almost a year of troubleshooting at Harvard to get everything working. She couldn’t use normal cultures to grow the organisms, so she worked with a team of medical engineers at Massachusetts General Hospital to devise microfluidic plates—little silicon wafers carved with minute channels and wells. Using these, she could isolate a single M. tuberculosis bacterium and keep it alive through at least five divisions. Meanwhile, she needed to write a lot of code from scratch to be able to analyze her images, and tweak her microscope cameras before they could focus reliably on the bacterial cells, which are about twenty times thinner than mammalian ones.

But when the work was done, the beginnings of an extraordinary story had emerged. “These two are sisters,” Aldridge said, gesturing to two long shapes, attached like sausage links. “You can see they’re not dividing at the same time, and they’re not growing from the middle or even at the same rate.” This was rare for bacteria, which typically grow and divide neatly, splitting into two identical daughter cells. Not tuberculosis, though: As Aldridge discovered, every time an M. tuberculosis cell divides, it makes two daughter cells that look and behave distinctly, even though they are genetically identical. One is always larger and faster-growing—the one that inherited the mother’s cell-growing machinery, or growth pole—while the other is invariably smaller, slower, and forced to make its own growing parts anew. These size differences become more pronounced as the cells progress through generations.

The asymmetric growth Aldridge discovered had profound clinical ramifications. In experiments, she found that the bigger, faster-growing cells—she named them accelerators—were sensitive to certain antibiotics that attacked their cell-wall making machinery, yet could withstand assaults from different antibiotics that targeted other cell operations. The little cells, which Aldridge called alternators, responded in the opposite pattern—perhaps because the alternators weren’t making cell walls at the same rate as their sisters, they were less vulnerable to the first type of antibiotics, but suffered more from other drugs.

The stunning findings finally answered the question of why tuberculosis was so hard to treat: Though genetically identical, different M. tuberculosis cells seem to have a built-in ability to tolerate different drug regimens. “We didn’t know that until we just started looking at the cells,” Aldridge explained.

“It totally makes sense—once you see it. But somebody has to think of it,” said Eric Rubin, M90, SK90, who leads a tuberculosis lab and chairs the Department of Immunology and Infectious Diseases at the Harvard T.H. Chan School of Public Health. That somebody, it turned out, was Bree Aldridge. Her insight, Rubin said, was “one of those paradigm- breaking things that instantly becomes a new paradigm.”

Before 2012, the Tufts’ Health Sciences campus didn’t have a tuberculosis researcher, and didn’t even have the right lab for it. But John Leong, chair of the Molecular Biology and Microbiology department, understood that modern technologies were likely to soon lead to major new discoveries in the TB field—and with Tufts’ reputation for microbiology excellence, it could be at the forefront. He even persuaded the medical school’s dean to promise several million dollars for a new Biosafety Level-3 lab, earmarked just for that disease.

All Leong needed next were scientists to populate the future facility. He’d heard about Aldridge from his friends, the two TB experts Rubin and Sassetti. “I didn’t know who Bree Aldridge was, but she was all about this computational approach and this single-cell analysis, which was very groundbreaking,” he said.

Before Leong had the chance to reach out to her, though, Aldridge spotted and applied for a pair of faculty openings at Tufts, in the departments of Molecular Biology and Microbiology at the medical school in Boston and Biomedical Engineering on the Medford/Somerville campus. Hoping to land one, she got them both. Her dual appointment was unprecedented for the departments. “We didn’t have a structure to do this, so it was much more complicated,” said Leong. But he anticipated that Aldridge would be a catalyst for more interaction between his and other groups at Tufts. “Interdisciplinary work is the wave of the future,” he said. “Tufts has got to do that—and has to do that well—in order to thrive.”

As soon as she arrived at Tufts in 2012, Aldridge began building her lab. She chose people less for their past expertise, and more for their future potential. “The work that we do is so specialized,” she explained. “I’m not looking for people with a certain skill set. I’m looking for people with the drive to learn new things to help solve a particular problem, who are willing to learn another discipline.”

Things were slow for the first year, and then in early 2013 she won a prestigious $50,000 fellowship grant from the Alfred P. Sloan Foundation. That fall, she followed it up by netting a $1.5 million NIH Director’s New Innovator Award, one of only 41 scientists nationwide to receive the honor. “That’s a major award,” Leong said. “It really puts an investment into investigators the NIH sees as being highly innovative and potentially growing into real leaders in their field.”

Aldridge has used her original time-lapse system as the foundation from which to study the odd TB microbes in greater detail. In one experiment, her lab studied how mycobacteria react to a single common antibiotic, rifampicin. They discovered the reaction is determined by a complex set of connected factors, such as how big and how old the cell is at the time a treatment starts, and where exactly the cell is in its growth cycle. And that’s just against one antibiotic—it gets more complicated.

In another study led by collaborators in Vietnam, Aldridge helped show that fighting the bacteria somehow caused some of the cell-size variation she had documented in her time-lapse studies. Clinical tuberculosis bacteria taken from patients showed a wide range of sizes—the more drug-resistant they were, and the worse the patient’s symptoms, the more the cells varied. When the Vietnam team cultured some of these TB-causing bacteria in conditions resembling the stressors of a human host, or treated them with antibiotics, the microbes’ lengths grew increasingly varied. And when the scientists exposed them to both conditions, the effect became even more pronounced.

What this means is that mycobacteria seem to change their size as a response, or maybe adaptation, to different stressors. Aldridge calls it a form of evolutionary “bet hedging.” The more that’s thrown at them, whether by the immune system or by antibiotics, the more the microbes divide into ever-varying lengths that somehow improve their ability to survive attack. What doesn’t kill tuberculosis literally makes it stronger.

Expanding on this has become a main focus of Aldridge’s lab, with postdocs and graduate students each tackling a piece of the mycobacterial response. One grad student, Michelle Logsdon, has already discovered how one of these species, M. smegmatis, controls its sizes in a different way than any other microbe studied. “I had my feet in two fields with this study,” Logsdon said. There was the relatively new mycobacterial growth field—familiar territory for the biologist—and the older niche field of cell-size control, which is dominated by physicists and mathematicians. At one point, Logsdon showed the mycobacteria’s bizarre abilities to their collaborators at Harvard’s School of Engineering and Applied Sciences. “They’re like, wait, bacteria can’t do that!” she recalled. Up to then, of course, they had mostly studied established model bacteria like E. coli. Logsdon is now trying to figure out if the tuberculosis bacterium is similarly unusual.

Elsewhere in the Aldridge lab, postdoc Trever Smith and graduate student Jonah Larkins-Ford are developing culture conditions that are more similar to the human host. “Since what the bacterium is introduced to and what environment it is in will influence the degree of tolerance,” Smith explained, “We’re trying to systematically go through what tuberculosis may be experiencing in different parts of its host.”

When Aldridge tested the antibiotic rifampicin, she compiled cell growth and cell-cycle parameters that she could use to make a computational filter of the different behavioral rules that the microbes seemed to follow. Working with Smith and Larkins-Ford, she’s creating a compendium of all the bacteria’s possible behaviors and abilities, adding each new one to the list as they uncover it. “If we’re going to design new interventions,” Aldridge said, “don’t we have to understand the basic function of the cells and how they achieve their crazy ability to tolerate stress?”

Last year, Aldridge came out with what others in the field call her most important work yet. It begins with the fact that treating TB requires taking multiple drugs at once. If scientists wanted to study how these drugs interacted with each other—if some of them negated another’s effects, say, or synergistically enhanced them—they would use a “checkerboard assay” with 384-well plates, each filled with increasing doses of drugs mixed together to see how they perform against the bacteria.

But no one has systematically done this for all of the countless possible TB drug interactions. That’s because while a single two-way drug analysis requires one such smartphone-size plate, a four-drug interaction requires 25 plates. “For a 10-way interaction, you’d have to stack them up as high as the Empire State Building,” Aldridge said. “And this is why people don’t measure drug interactions at high order.”

Aldridge and a visiting scholar in her lab, Murat Cokol, however, recognized that the most important information a researcher needs is simply whether the drugs interact positively or negatively. For that, she didn’t need data from every possible combination, just a very few key ones. So last fall, in collaboration with Cokol and others at Harvard and Sabanci universities, she helped develop a mathematical model that lets her extrapolate from a subset of those most important combinations. “It’s not unfair to call it quick and dirty,” she said of the new method. “It’s just that in order to be quick and dirty, we’re really careful about our experiments.”

In her first paper on the method, which she has named DiaMOND for “Diagonal Measurement Of N-way Drug interactions,” Aldridge easily identified new synergistic combinations of two, three, and four different TB drugs. “We’ve measured up to 10-way interactions now,” she said. “You can keep piling this up.” She has already started expanding this effort to create new tuberculosis drug combinations for preclinical testing with the help of a nearly $1.3 million grant awarded by the Bill & Melinda Gates Foundation this March.

“It takes somebody thinking about the complexity of the problem from a quantitative perspective to see the solution,” said Sarah Fortune, Aldridge’s postdoc advisor at the Harvard Chan School. “And I think it speaks to a larger problem, which is how do you develop combination therapy for anything, even cancer?”

At Tufts, Leong agrees. “This is really exciting,” he said. “It has the potential to make the difference in how we approach the development of drug regimens. And that could be in TB, or any number of drug-resistant pathogens.”

Aldridge is already working with other researchers at Tufts University School of Medicine to expand the DiaMOND method, including Joan Mecsas and Ralph Isberg, who study common hospital-acquired pathogens caused by bacteria such as Klebsiella and Acinetobacter. The mathematical model, Aldridge said, “is plug-and-play.”

Those are far from Aldridge’s only cross-department collaborations. She has a project in the works with biomedical engineer Fiorenzo Omenetto to develop even better testing and analytic technologies to study TB, and is in discussions with bioengineering chair David Kaplan about developing realistic 3D models of human lung tissue to study TB and other lung conditions, such as fibrosis.

Just as Leong had hoped, Aldridge’s interdisciplinary approach has caught on throughout the microbiology and bioengineering departments. “When she came, we had virtually no interactions with the biomedical engineering department,” Leong said. But as soon as she arrived, she got the two groups talking, sharing their research, finding new ways to blend their expertise. The departments ultimately netted a joint almost-$8 million grant from the NIH’s National Institute of Allergy and Infectious Diseases to establish a research center dedicated to the study of dangerous bacteria in bioengineered models of the human intestine.

“Having faculty who are trained with one foot in engineering and one in biology is fantastic, because they understand the language in both places, and that can synergize this dual approach to research,” said Kaplan, who is the grant’s co-primary investigator with Isberg. Aldridge, he added, “was a catalyst, a translator—she helped to bring the two departments together.”

Back in her lab, Aldridge plans to combine her two primary research threads, folding some of her work on the TB bacteria’s response to immune and antibiotic stressors into her DiaMOND drug-interaction tool. “The principle of this method is that as long as each stressor has a dose-response, then you can use DiaMOND to efficiently measure how it all behaves in combination,” Aldridge said. And that might eventually lead to shorter and more effective drug strategies that save lives.

Right now, everything is still very preliminary, but Smith and Larkins-Ford are already attempting to translate the more immune system-mimicking cultures they’re creating into DiaMOND. “It’s like all the tools we use in the lab are systematically meshing together in this beautiful way,” Smith said.

Sometimes, bioengineers define their field as harnessing biology to engineer new systems. But Aldridge flips that on its head, using math and pattern-finding technologies to unravel biological puzzles. “That’s the cool thing about these tools,” Aldridge said. “You can have a huge problem that seems insurmountable when suddenly, you see a pattern.”

“And then we can use that knowledge to cut through the complexity and understand biology in a way that we couldn’t see before.”