Creating Biomedical Technologies to Improve Health


Science Highlight: January 15, 2012

A Tale of Two Livers - Developing an Innovative Drug Screening Tool

New drugs undergo rigorous safety and efficacy testing in animals before they are tried in people. However, due to differences in the ways humans and animals process drugs in the liver, animal models cannot predict side effects that are unique to humans. To bring animal models closer to mirroring human physiology, researchers have recently been developing a humanized mouse model that harbors human liver tissue. These humanized animal models provide insight into how disease processes and new therapies might behave in people.

Reconstructing Human Liver Tissue

Humanized mouse models are difficult to produce and maintain. For survival, transplanted human liver cells must establish intricate signaling molecule exchanges with mouse tissues, as well as evade attack by the mouse’s immune system. Massachusetts Institute of Technology (MIT) Professor Sangeeta N. Bhatia, (who teaches in both the Health Sciences and Technology and the Electrical Engineering and Computer Science Programs), theorized that integrating human liver cells with a supportive and protective microenvironment, prior to their implantation, would improve the survival and the functioning of humanized liver tissue transplanted into a normal mouse.

A deep understanding of cell-cell interactions, along with a sophisticated level of tissue engineering, was required to generate the humanized livers for implantation. In engineered liver tissue, the microenvironment consists of a polymer scaffold and supporting cells. After testing several combinations of ingredients, Bhatia’s research team uncovered a recipe for tissue they dubbed human ectopic artificial liver (HEAL). In the first step, they grow (in vitro) human hepatocytes, the predominant liver cell type, along with mouse connective tissue cells called fibroblasts. After one week, the hepatocyte/fibroblast cell clusters are mixed with human endothelial cells (cells that line the inner surface of blood vessels), and the combination is encapsulated within photo responsive polyethylene glycol diacrylate (PEG-DA) polymer scaffolds. When exposed to light, PEG-DA polymer chains interlink to form a mesh. “You shine a light on it, and it entraps the cells,” says Bhatia. The PEG-DA scaffold contains biologically active molecules that facilitate interactions between cells in three-dimensional space. Fibroblasts provide cues to stabilize hepatocytes while they grow in vitro, and endothelial cells secrete signals needed for the recruitment of blood vessels that are critical to the survival of the construct after implantation.

The liver processes drugs similar to how the gut processes food, by breaking the drugs down into smaller components called metabolites (and some metabolites may be very harmful to people). During the drug breakdown process, a mouse liver might not produce all the same metabolites that a human liver would. “You don’t want to see a [toxic] drug metabolite for the first time in humans, one that you hadn’t seen in animal testing,” says Bhatia. The resulting innovation came about by challenging the current thought that one needed to replace the mouse liver with a human one. The realization that it was enough to implant a human liver that could function side-by-side with the mouse liver was the revelation. Alice Chen, a graduate student in Bhatia’s MIT lab, turned this idea into reality by grafting these human liver “organoids” into normal mice to see whether the new tissues could produce human-specific drug metabolites. Bhatia’s research team determined that HEALs express relevant human drug-metabolizing enzymes, proving their potential utility for drug metabolism studies. Subsequent experiments confirmed that HEAL-humanized mice could produce human-specific drug breakdown products that do not appear in regular mice. HEAL-humanized mice have also proven useful for probing drug-drug interactions, a critical determinant for drug safety and efficacy. Such assessments could be used to prevent harmful and/or ineffective drugs from advancing to clinical testing—thereby saving time and money while also reducing unnecessary human suffering.

Tweaking the Recipe

As they already contain a supportive microenvironment, HEALs can be implanted in the abdominal cavity or under the skin of mice, and they function independently of the existing mouse liver. Within a week from implantation, HEALs connect with mouse vasculature and start producing human liver proteins. The same process takes 2–6 months in other humanized mouse models, where transplanted cells need to travel to the mouse liver and “set up shop.” In these other models, the mouse liver must be injured to make room for transplanted cells, and mice must be immunosuppressed to prevent human cell rejection. Ideally, “to get clean results for certain human metabolites, you need to implant HEALs into different mouse models, including those with intact immune systems,” says Bhatia. Because the polymer can act as a barrier to attacks from the immune system, HEALs could potentially be used in animals with normal immunity.

The initial results in HEAL-humanized mice are very promising, yet there is room for improvement. Bhatia’s collaborator, Christopher Chen, Skirkanich Professor of Innovation in Bioengineering at the University of Pennsylvania, is exploring ways to enhance HEAL vascularization. “Our joint effort is looking at how biomaterials, cell organization, and different types of cells might support the hepatocytes to accelerate integration of implanted tissue and extend implant lifetime,” he says. This research will also address challenges related to making larger implants for therapeutic applications in people. “We’re looking at different adhesion molecules and changing how the cells are organized in the implant.”

A: To the naked eye, a HEAL looks like a pink contact lens. Approximately 4/5 of an inch in diameter, the implant comprises half a million liver cells. B: Growth of blood vessels (red) enables implanted HEALs to grow and function in the mouse. This miniature human liver was removed from a HEAL-humanized mouse.

A: To the naked eye, a HEAL looks like a pink contact lens. Approximately 4/5 of an inch in diameter, the implant comprises half a million liver cells. B: Growth of blood vessels (red) enables implanted HEALs to grow and function in the mouse. This miniature human liver was removed from a HEAL-humanized mouse.

Bhatia is also developing HEAL mouse models for hepatitis C, malaria, and other human pathogens for which no animal models are available. These new models would allow researchers to grow human viruses—that normally don’t infect mice—in a human liver setting to study human drug responses.

In the future, HEALs created from different donor cells could be used to study drug effects in people who are susceptible to drug-induced liver injury, as well as people who metabolize drugs at different rates (so-called slow versus rapid metabolizers). A person’s metabolism rate can affect drug efficacy and safety. “Slow metabolism can increase exposure to the drug and increase toxic side effects. Fast metabolism can cause rapid clearance and reduce efficacy,” explains Bhatia. To study rare liver diseases such as alpha-1-antitrypsin disease, Bhatia is constructing HEALs from donor cells that can be harvested from blood or skin and converted to hepatocyte-like cells in the lab. In addition, this new technique offers the opportunity to evaluate drug-drug interaction in a living organism, something that is not possible in static in vitro systems.

Chen and Bhatia’s research is laying the groundwork for the option to construct an implantable artificial liver as an alternative to whole-organ transplantation for people with liver failure. The PEG-diacrylate skeleton is suitable for organ printing—that is, assembling organs layer by layer. “Some of the technologies that we are developing will hopefully even translate to engineering other organs,” says Chen.

This work is supported in part by the National Institute of Biomedical Imaging and Bioengineering; the National Institute of Diabetes, Digestive, and Kidney Diseases; the National Cancer Institute; Howard Hughes Medical Institute; and the National Defense Science and Engineering; and National Science Foundation Graduate Research Programs.


Chen AA, Thomas DK, Ong LL, Schwartz RE, Golub TR, Bhatia SN. Humanized mice with ectopic artificial liver tissues. Proc Natl Acad Sci U S A. 2011 Jul 19;108(29):11842-7.

Underhill GH, Chen AA, Albrecht DR, Bhatia SN. Assessment of hepatocellular function within PEG hydrogels. Biomaterials. 2007 Jan;28(2):256-70.

Program Area: 
Health Terms: 
Liver Disease,
Tissue Engineering/Regenerative Medicine