Enhancing Drug Responsiveness with Synthetic Liver Microenvironments
Humanized mice with artificial liver tissue
Despite advances in growing human cells, like liver hepatocytes, in culture, there are many aspects of biology and medicine that cannot be mimicked in a dish. Another approach that scientists is to grow human cells in an animal host, so that the transplanted human cells behave more like they would in a person. These so-called "humanized" mice offer a window into important aspects of human disease such as predicting dangerous drug toxicities. The best available methods for liver humanization rely on transplanting cells into mice with liver injuries, which allows the human hepatocytes to grow beside mouse cells in the animal's liver. However, these models only work well if the animals do not mount any immune responses, else they would reject the human cells, and require that the liver is already damaged in some way. Most challenging is that the success rate of the human cell engraftment is inconsistent and often not very extensive. We sought to develop a humanized mouse model based on a simple method of implanting small, consistent quality human "micro-livers" that are easily made using tissue engineering techniques. We call them HEALs (human ectopic artificial livers), and when we implant them into mice we observe humanized liver functions in the blood of the hosts for weeks. HEAL-bearing mice could predict the disproportionate metabolism and toxicity of "major" human metabolites (Chen et al., PNAS 2011).
Human ectopic artificial livers for humanized mice. (A) Fabrication and implantation of HEALs for humanizing mice. (B) Modification of PEG-DA hydrogels with RGDS peptide and co-culture of hepatocytes (HEP) with fibroblasts (FIB) and endothelial cells (LEC) improves tissue function. (C) Luciferized HEALs persist in vivo, secrete human albumin into mouse circulation, and identify major human metabolites in mice. (D) Microstructural placement of endothelial cells modulates hepatic tissue function. (E) Endothelial cell patterning templates ingrowth of perfused chimeric capillaries. HEALs with EC cords can support human hepatocyte function in vivo.
Because we had previously demonstrated the importance of 2D tissue architecture (i.e. organization through micropatterning) in hepatic function, we next explored the role of tissue architecture in 3D. We developed micromolding methods to control the organization of clusters of hepatocyte and how these groups localized relative to blood vessel cells called endothelium inside the 3D hydrogel. Using this method, we found that the functions performed by hepatocytes depend on local interactions with other cell types, and that we could optimize the structures maintain human liver cell functions over long periods inside host mice (Stevens et al., Nat Comm 2013). To investigate the importance of the organization of the endothelial vessel cells, we collaborated with Dr. Christopher Chen (UPenn – now at BU) to to pre-organize different patterns of 'cords' of vascular cells. HEALs containing cords and hepatocyte clusters engrafted more rapidly than first-generation HEALS, and resulted in organized networks of blood vessels that connected with the host mouse blood flow (Baranski et al., PNAS 2013; Miller et al., Nat Mat 2012).