Humanized mouse livers offer numerous drug testing possibilities

CAMBRIDGE, Mass.—Big help may be coming for pharmaceutical companies in the form of some very small livers, as exemplified in a recent paper published by the MassachusettsInstitute of Technology (MIT). In that paper, researchers detail a new technology that could helpresearchers know exactly how a drug will affect humans before it goes intoclinical trials.

Clinical trials are a huge investment forpharmaceutical companies, one that banks on the drug or drugs being testedworking exactly as hoped. However, some drugs end up displaying side effectsthat can be damaging or even fatal to study participants—side effects thatdevelopers couldn't predict before the trials. This can lead to a trial beingscrapped early, costing millions of dollars and setting a drug back by years.

Now, though, Alice Chen, a graduate student in the MIT-Harvard Divisionof Health Sciences and Technology (HST), has come up with a way to grow humanlivers inside of mice to produce "humanized" mouse livers that process andmetabolize drugs the same way human livers do. While mice have been used for yearsin research, they are less effective in pharmaceutical testing since mouselivers process drugs differently than human livers. The discovery offers theability to study drugs that might harm the liver before they get into clinicalstudies, as well as the human liver's reaction to diseases such as malaria andhepatitis.

Chen works in the laboratory of Sangeeta Bhatia, the Johnand Dorothy Wilson Professor of HST and Electrical Engineering and ComputerScience. Chen's findings were published in the Proceedings of the NationalAcademy of Sciences (PNAS), and Bhatia, a member of MIT's David H. KochInstitute for Integrative Cancer Research, is senior author of the paper.

The idea of humanized livers, says Chen, came from herdesire to help advance the field.

"It sort of dawned on us one day that this would be useful,just having human liver tissues," says Chen.

Chen and Bhatia's method of creating the humanized mouselivers consists of a tissue scaffold that includes nutrients and supportivecells, which preserve liver cells after they are removed from the body. Thescaffold is the size and shape of a contact lens, with a similar texture, andis implanted into a mouse's abdominal cavity. The gel that the scaffold is madeof also doubles as a barrier, keeping the mouse's immune system from rejectingthe implant.

"There are methods of humanizing a liver in mice that havebeen published before that are very different from what we did," explains Chen."In previous methods, researchers have really elegantly injected liver cellsinto mice that have a genetic liver injury, so it causes the liver cells totravel to that injured mouse liver, and then it clamps there and startsgrowing."

The issue with that method, however, is that it drasticallylimits which mice can be used for testing, as researchers would have to breedadditional generations to have enough subjects to use in testing. With Chen'smethod, up to 50 mice can be implanted with scaffolds in a day, and it takesonly about a week for the implanted tissue to fully integrate itself into themice. In the paper's abstract, the researchers detail how the implants, orhuman ectopic artificial livers (HEALs), "stabilize the function ofcryopreserved primary human hepatocytes through juxtacrine and paracrine signalsin polymeric scaffolds." Unlike the current method, the HEALs can be introducedin mice without compromised livers, and mice implanted with HEALs "exhibithumanized liver functions persistent for weeks, including synthesis of humanproteins, human drug metabolism, drug-drug interaction and drug-induced liverinjury."

"We worked really hard to make sure the human livers werejust working outside of the mouse first, that was a really big, important firststep that took actually several years," says Chen, adding that there were somesurprises when they studied how well the implants integrated into the mice. "Wehad cell types in our device that we knew would secrete factors that would workthrough the mouse blood vessels in our samples. We didn't know that the vesselswould go through the implant, we thought they might just come to the surface ofthe implant."

The liver tissue successfully integrates into the mouse'scirculation system, allowing drugs to reach it and proteins produced by theliver to enter the bloodstream. Though the mice maintain their own livers, theresearchers are able to distinguish between the responses of mouse and humanliver tissue. Administering dosages of coumarin and debrisoquine proved thatthe mice metabolized the drugs into the byproducts generated by normal humanlivers.

Moving forward, Chen says researchers will be studying howthe humanized livers process other drugs whose metabolites are known, whichwill help them gauge the implants' reliability in terms of testing new drugs inorder to predict how they will be processed in humans. In addition, theresearchers will also work towards replicating the implants at even smallersizes in hopes that potentially hundreds could be implanted in one mouse. If itworks, it could drastically reduce how many mice would be needed for studies.

"We're also interested in looking at drug pairs, being ableto create a battery of mice that have human livers that you can start reallytesting different combinations of drugs," Chen adds. "We're interested inlooking at drug interactions, potentially dangerous drug interactions thatwould be hard to test prior to clinical trials and marketing."

The paper, entitled "Humanized mice with ectopic artificialliver tissues," was published in the July 11 issue of PNAS. In addition to Chenand Bhatia, the paper had four other authors: David K. Thomas of the BroadInstitute of MIT and Harvard and the Division of Adult Palliative Care,Dana-Farber Cancer Institute and Harvard Medical School; Luvena L. Ong of theHarvard-MIT Division of Health Sciences and Technology; Robert E. Schwartz ofthe David H. Koch Institute and the Division of Medicine at Brigham and Women'sHospital; and Todd R. Golub of the Broad Institute of MIT and Harvard and theHoward Hughes Medical Institute at MIT.