Linda Griffith - Marrying tissue engineering with systems biology

Linda Griffith combines in vitro models with deep molecular analysis to accelerate drug discovery.
Eric Bender | MIT Industrial Liaison Program
January 31, 2014

“Together, we’re building interactive organs,” says Linda Griffith, professor of biological and mechanical engineering. That’s the goal of the Defense Advanced Research Projects Agency (DARPA) research program known formally as the Microphysiological Systems program and informally as the “body on a chip” program.

“DARPA had a vision to replicate all 10 human physiological systems on a single research platform to allow the researcher to look at crosstalk between the systems, such as the liver with the gut or with reproductive systems,” Griffith says. She directs the Barrier-Immune-Organ: MIcrophysiology, Microenvironment Engineered TIssue Construct Systems (BIO-MIMETICS) program, one effort under the DARPA program. BIO-MIMETICS has numerous collaborators at MIT as well as the Charles Stark Draper Laboratory, the University of Pittsburgh, Northeastern University, and Zyoxel of Oxford, United Kingdom.

Griffith’s work builds on her lab’s earlier development of a three-dimensional “bioreactor” model of the human liver, now being commercialized by Zyoxel for use in drug discovery.

Additionally, her latest research draws heavily on systems biology, which investigates how cells integrate information across multiple scales, including genes, RNA, proteins, protein activity states, and metabolites. “Now we can make high-information-content measurements on all of these scales of information flow in a cell,” Griffith says. The research, which includes several MIT biological engineering faculty members, focuses particularly on protein activity states and the effects of those states — for example, how a cell responds to a molecular growth factor — using mathematics to model cell behavior.

“We view the merging of tissue engineering and systems biology as an incredibly powerful marriage for the future of the whole drug development process, everywhere along the pipeline up to clinical trials,” Griffith says. “This will let us relate the kinds of measurements we can make in a patient to the kinds of measurements we can make in our in vitro model. It’s about more than if the cells metabolize the drug; it’s about actually understanding how the drug works.”

Modeling liver inflammation and metastatic cancer
One “body on a chip” program that Griffith leads examines interactions between models of the gut and the liver. “We’re building a very detailed model of the gut/liver interaction that lets us actually start to put real patient gut bacteria in a model of the human intestine and have it interact with the liver,” she says.
“All the blood in your intestine immediately goes to the liver, which regulates your metabolism, but your gut’s also filled with microbes, and little pieces of microbes leak across the gut wall all the time and interact with the immune system in the liver,” she explains. “If you get a gastrointestinal disease or take a drug that changes the gut permeability, now all of a sudden the liver can see a lot more bacterial products than it’s used to, and it can get inflamed. That may be okay, but if you’re taking a drug or you have some kind of stress, you may harm your liver.”

In another closely related project, funded by the National Institutes of Health, Griffith collaborates with colleagues at MIT and the University of Pittsburgh to investigate metastatic triple-negative breast cancer tumors placed in a 3-D liver bioreactor.

“I myself had triple-negative breast cancer, and I came to appreciate how hard it is to develop new cancer drugs to treat patients,” Griffith says. “We don’t know why chemotherapies fail in a very significant percentage of patients with triple-negative breast cancer.”

Those patients typically die of metastasized tumors that have spread to the liver and other organs. Tiny metastatic tumors in these organs may lie dormant for years, and the combination of tissue engineering with systems biology provides new opportunities to study what triggers the tumors’ activation.

“You can put cancer cells in our liver bioreactors and get them to quiet down, just as one of these silent micrometastases might do in a patient,” Griffith says. “Now we have a human model that lets us study what stimulates those cancer cells to die or proliferate.”

These highly aggressive cancer cells typically keep growing when placed in normal cell cultures, but Griffith and her research partners have created what they think is the first evidence that cells can be kept inactive in vitro. “That’s a huge advance, because now we can start to bring them in and out of that quiescent state in tissue, and really have a model for them,” she says.

Advances in endometriosis
Griffith and her colleagues also take a systems-biology approach in an investigation of endometriosis, a painful and sometimes debilitating condition in which pieces of the uterine lining migrate out of the uterus and grow in the abdominal cavity. The disease afflicts up to 10 percent of women, often takes years to diagnose, and is typically treated only with surgery.

Griffith has collaborated with MIT professor Douglas Lauffenburger (a leader in systems biology research) and surgeon Keith Isaacson of Newton-Wellesley Hospital on a clinical study of inflammation mechanisms in endometriosis.

Analyzing patient samples for levels of 50 cytokines involved in immune cell signaling, the researchers found that in one group of patients the levels of 13 cytokines were elevated together, suggesting that the cytokines were part of a common molecular network. Further study revealed details of how the inflammation process is driven by macrophage immune cells. Moreover, the findings pointed to a molecular pathway that hadn’t previously been implicated in endometriosis and that can be inhibited in cell culture by existing small-molecule agents.

While this research is still in its early stages, its promise of drug treatment for endometriosis has provoked excitement among gynecologists. “We’ve galvanized a whole international community to think about the disease in a new way,” Griffith says.

More broadly, “there are tremendous opportunities for this systems-biology approach in women’s reproductive health,” Griffith says. She is also bringing her tissue-engineering expertise to the field; the “body on a chip” program that she directs will incorporate tissue models addressing aspects of women’s reproductive health.

Griffith directs the MIT Center for Gynepathology Research, established in 2009 to help meet what she calls “a huge unmet need in women’s reproductive health” — finding better treatments for conditions such as endometriosis, fibroids, and adenomyosis that have seen relatively little federal research funding and few advances in recent years.

The Center brings together a broad group of MIT researchers with gynecological surgeons and industry partners. “We’re working very closely with wonderful advisors in industry who are cheering us on, because until they have models of efficacy and can unravel a new research direction it’s very hard for companies to make a commitment to these diseases,” Griffith says.

Partnering with pharmaceuticals
The Center is just one framework in which Griffith works with numerous pharmaceutical partners.

“Drug companies have a pipeline, and we’re adding to the tools in their pipeline,” she says. “What we’re building is not replacing regular cell culture or animals, it’s adding the ability to bring part of a clinical trial further up in the lab. So before you go into people, you’re actually trying to bring the people to the lab, in a miniature form, and test things on them.”

“There really is a huge interest in how do we bridge the in vitro-in vivo correlations for drug efficacy and toxicity,” she says. “And we now appreciate that it’s very complicated by individual patient situations, such as taking another drug or having an underlying stressor such as obesity.”

“There are many different facets of patient states that we need to capture, and we need to do it in an efficient way,” Griffith adds. “We communicate very frequently with our industry advisors about this topic, and it’s a very interesting two-way flow of information. We send people to their labs so we can see how Amgen, Novartis, Pfizer, or Sanofi would actually use a model. We don’t want to make it too complicated for them to use in a meaningful way.”